Skip to main content

Full text of "The Denatured Novel"

See other formats


gj<OU 158422?^ 

>& ~ %7) 



Call No. ***' *[V1 *&* Accession Ncx Cf /O frfl f~ 

Author \Ta#\ tVc*^r**&j ^V . 

Title ^j C-TrxcV'TUAi ^1 *V\0 \JoJL I ^IXlpO 

Tliis book should be returned on or before the date last marked below. 

The Denatured Novel 



THE CONSCIOUS VOICE (ed. with C. H. Watts) 




Albert Van Nostrand 




Portions of this book have appeared in: American Quarterly (Summer, 
1956), and The English Journal (January, 1959, and February, 1959) . 
Passages quoted herein from I Wanted to Write, by Kenneth Roberts, 
copyright 1949 by Kenneth Roberts and Anna M. Roberts, are re- 
printed by permission of Doubleday & Co., Inc.; from From Main 
Street to Stockholm: Letters of Sinclair Lewis, selected by Harrison 
Smith, copyright, 1952, by Melville H. Cane and Pincus Berner, execu- 
tors of the estate of Sinclair Lewis, are used by permission of Harcourt, 
Brace & Co., Inc.; from North of Grand Central, by John P. Marquand, 
by permission of the publisher, Little, Brown & Co.; from Past All Dis- 
honor, by James M. Cain, by permission of the publisher, Alfred A. 
Knopf, Inc.; from Hollywood: The Dream Factory, by Hortense Pow- 
dermaker, by permission of the author. 





./m. GRANT from the John Simon Guggenheim Memorial Foundation 
made this book possible. With this grant, during a sabbatical leave 
from teaching, I was able to devote my time and energies for an entire 
year to the systematic development of an idea. For this I thank the 
Trustees of the Guggenheim Foundation, and its Secretary General, 
Henry Allen Moe. 

The idea of the denatured novel encompasses both literary theory 
and economic cause, and my research has had to be broad as well as 
intense. To this end, many persons have shared with me their knowl- 
edge, their opinions often their disagreements and their specula- 
tions. I particularly remember the help of Joseph Barnes, Knox Burger, 
William Charvat, Alan Collins, Saxe Commins, Paul S. Eriksson, 
Hugh MacNair Kahler, Freeman Lewis, William Miller, Charles Nei- 
der, John M. Pickering and Victor Weybright Through various stages 
of manuscript, the idea of the book became more precise as a result of 
the demanding responses of Jess B. Bessinger, Jr., Peter H. Davison, 
Henry Klinger, Louise Lawrence, Robert Rieman, Judith Rothschild 
and Aaron Sussman. 

During six years of research and writing I have also, necessarily, been 
a student, and I have been blessed with wise teachers. I owe a great 
debt to Howard Mumford Jones, a teacher who understands the whole 
endeavor of publishing, and to Monroe Stearns, an editor who is first 
a teacher. 

My thanks are not intended to assign responsibility for the judg- 
ments in this book. The judgments are mine. But all of these persons 
have helped me to evaluate a significant subject for the community to 
share which is the aim of scholarship; and for their help I am grateful. 

Brown University 












INDEX 217 

The Denatured Novel 

Chapter I 

Art and Artifact 

BEFORE the history of the American novel began, fiction was already a 
commercial product in the Colonies. The novel was a form of enter- 
tainment already fortified with theories of art when it was imported 
into the Colonies in the eighteenth century, and it became domesti- 
cated in the new republic because a great many persons who were will- 
ing to buy its entertainment also wanted a certain record of themselves 
and a certain assurance of their possibilities. Accommodating the en- 
tertainment to the practical need for this self-assurance quickly became 
a business matter. Since popular tastes were crystallized before profes- 
sional writing was possible in this nation, the American domestication 
of the novel necessarily involved the abrasions of art and commodity. 

There never was any doubt about the commercial patronage of the 
new literature, but there were plenty of misgivings. The Reverend 
Samuel Miller summed these up, in 1803, in A Brief Retrospect of the 
Eighteenth Century, concluding his two-volume survey on a note of 
alarm. "Booksellers have become the great patrons of literature/ 1 he 
warned. The "SPIRIT OF TRADE" prevailed and authors now wrote 
"in accommodation to the publictaste, however depraved, with a view 
to the most advantageous sale." These were the auspices of the Ameri- 
can novel. 

But, in a legal sense, the domestication of fiction did not occur until 
1891, when Congress finally passed the first International Copyright 
Act, which forced American publishers to pay foreign authors for the 
privilege and profit of reprinting and selling their books in the 

1 4 The Denatured Novel 

United States. Tardy and in many ways inadequate though it was, the 
Copyright Act was the best compromise that a few publishers, trade 
journalists and legislators could achieve after a half-century of lobbying. 

The purpose of the Congress had been to protect American book 
manufacturers. So far as the legislators were concerned, literature was 
something of a by-product. But the new Copyright Act inevitably had 
literary consequences. By prohibiting piracy, the law made it equitable 
for the first time for American authors to be paid for their efforts. Since 
piracy had involved the reprinting of British novels, the law effectively 
decreed a domestic market for American novels. 

The new market for fiction after 1891 inherited the old contradic- 
tion. The venerable literary form still had to satisfy the fashions in en- 
tertainment, which meant gratifying certain popular beliefs, but this 
old contradiction of manufacturing art took on a new economic ur- 
gency. An expanding market for literature brought rising costs, so in 
order to sell what they manufactured, publishers engaged vigorously in 

The merchandising of literature was both the symptom and the re- 
sult of the changes which occurred in publishing during the 1890*5, 
and made the industry substantially what it is today. These economic 
changes, which resulted in the new fashion of marketing, were efforts 
to solve the industry's compelling problem of over-production. Sud- 
denly, publishers had a staggering quantity of books, with no assurance 
of their getting to their prospective purchasers. 

Back in 1804, the year after Samuel Miller had concluded his back- 
ward glance, A Catalogue of All the Books Printed in the United States 
listed a total of 1,338 titles. But in 1901, ten years after the Copyright 
Act, there were more than 150,000 titles in print in the United States, 
to which publishers were adding about 8,000 new titles a year. But the 
industry had no system of distribution which could keep pace with this 
volume. What the Reverend Miller had called the "unprecedented 
multiplication of books," nearly a century before, had by now gone 
beyond even the publishers. 

By 1891, publishers had abandoned the practice of auctions, by 
which they had always sold their wares to the trade, in favor of a net- 
price system of wholesaling and retailing. But they soon discovered 

Art and Artifact 15 

that they still had no control over the price of a book or the promotion 
of it to a putative customer. After John Wanamaker's department store 
had proved how successfully books could be sold as sidelines to other 
consumer products, dry-goods outlets, bazaars, and even drug stores 
followed suit, with the result that the industry's trade paper, Publishers 
Weekly, expressed dismay over the indignity of this "bazaar spirit" in 
the book trade, and feared for its effects on literature. 

Books suffered far worse than indignity by this kind of selling. The 
easiest way for the retailer to make money was to stock only a few titles 
and to sell them in volume, a practice that concentrated the consumer's 
attention on only a fraction of the titles publishers had made available. 
It penalized the great majority of books, and aggravated the problem of 

With this sort of merchandising the department stores could under- 
sell the bookstores which had to maintain more expensive inventories. 
No reprisal was possible. When the American Publishers' Association 
obtained court injunctions against the R. H. Macy Co., in 1901, and 
began a long war of attrition, Macy filed suit for damages against this 
monopoly, and won. The Supreme Court decision of this case, ^1914, 
protected that practice of bookselling which denied an outlet to most 
of the books published. 

Ironically the entire trade's active promotion of books during the 
iSgo's accomplished only the same limiting emphasis. Publishers began 
to advertise their books, and circulate review copies to magazines and 
newspapers, in hopes of gathering opinions which they could publicize; 
new journals carried the advertising and supplied the publicity. The 
most influential of these were the Bookman, founded in 1895, and the 
weekly Book Supplement of The New York Times, which appeared a 
year later; but these could select only a few of the titles for special at- 
tention and merely listed the others. 

A cataloging system of one of these journals, in fact, soon became 
the booktrade's favorite and most limiting promotional device: the 
best-seller list. At the back of its first issue Bookman listed retailers' 
reports of the six books currently leading in sales in sixteen selected 
cities; and in November 1897 the same magazine compiled the first 
national survey of "Best Selling Books." To all intents the trade en^ 

1 6 The Denatured Novel 

dorsed this summary bookkeeping in 1911, when Publishers' Weekly 
printed its first annual "Best Selling Consensus." And this channeling 
encouraged the folly of best-sellerism. 

What Publishers' Weekly called the "bazaar spirit" affected the en- 
tire book trade. Subsequently the sales value of literary property in 
the cataloging, retailing and advertising of books was systematically 
exploited to form public opinion. Best-sellerism affected even the edi- 
torial part of publishing. The domestication of literature by the Inter- 
national Copyright Act had enabled the book trade to publish Ameri- 
can novels, but what the trade promoted was not an art form but an 

Meanwhile, domestication of another sort in the preferences of 
American readers and book buyers had already developed. Popular 
opinion favored English novels of adventure and sentimentality. The 
first novel published in the Colonies was Benjamin Franklin's reprint 
of Samuel Richardson's Pamela; or, Virtue Rewarded, in 1744; and 
English novels were imported in considerable quantity even during and 
after the Revolution. British literary opinion fashioned some of the 
taste for the novels of Richardson and Laurence Sterne (then Scott, 
and later Dickens, Bulwer and Thackeray). But the word was really, 
and literally, spread by American booksellers who enterprisingly re- 
printed every novel they could possibly steal from the British pub- 

American writers complained that the demand for this fiction 
stunted native literary development. James Fenimore Cooper's two- 
volume description of his homeland for European readers, Notions of 
the Americans (1828), contains only a few paragraphs about American 
writing, and these begin dismally: "As respects American authorship, 
there is not much to be said." Cooper's book only emphasizes the dis- 
crepancy between America's literary dependence and its political in- 
dependence, for which the author blamed the absence of international 
copyright and the piracy of the American booksellers. 

Cooper's contemporary, John Neal, also complained about the copy- 
right situation, but he went further: ". . . if we would write for noth- 
ing," Neal interrupted his novel, Randolph (1823) to say, "and give 
the copyright of a novel, for instance^ to a publisher, it would still be a 

Art and Artifact 17 

perilous adventure to him." He was sure no literary form wielded so 
much influence on society as the novel, but novel writing in America 
had been relinquished by men of genius and given over to women 
and children. 

The taste of American book readers, Neal charged, was dictated by 
British book reviewers. He mocked the public which had waited twenty 
years to read the novels of Charles Brockden Brown, "till he had under- 
gone an apotheosis at London." Poe and William Gilmore Simms and 
a dozen other writers lamented this same sad state of affairs, but to no 
effect. Until the middle of the nineteenth century novels which ex- 
ploited the inherited taste for English romances were what American 
publishers wanted. 

Virtually every historical account of early American fiction begins 
with an inventory of the curses and benedictions over the novel, of its 
evils and its virtues. Fiction was suspect so long as the Neo-Classical 
fashion of distinguishing between entertainment and instruction held 
sway, as well as the Puritan fashion of deploring this sort of entertain- 
ment. But the moralists might as well have been talking to themselves; 
the booksellers kept right on grinding out their sentimental and 
wicked novels. 

Profit and pietism, in fact, kept each other's company. The nine- 
teenth-century magazines contained the same wicked fiction their edi- 
torials denounced. The advertisements of booksellers protested that 
novels were not their primary stock in trade, but they could scarcely 
stock enough of them to meet the demand. Even in the novels them- 
selves sentimental heroines were forbidden to read novels. All told, 
the reader could wallow in fiction's pleasurable vices and save his face 
at the same time. 

The charges against fiction usually specify its wickedness (novels 
were blatant lies ) and its wastefulness. To the charge of lying, the titles 
of the novels themselves replied. Two of the earliest of these notorious 
fictions spoke for dozens of others: Susanna Rowson's Charlotte Tem- 
ple; A Tale of Truth ( 1791; 1794), and Hannah Foster's The Coquette: 
a Novel Founded on Fact (1797). A shrewder shot at a wider target 
was Isaac Mitchell's The Asylum, or Alonzo and Melissa, an American 
Tale Founded on Fact (1811). 

1 8 The Denatured Novel 

And the novels carried other armament. A rebuttal to the charge of 
fiction's immorality usually appeared in a preface, sometimes called an 
"Advertisement" or a "Foreword," an "Introduction," "A Note to the 
Reader," "From the Author to the Reader," "Introductory Epistle" or 
"A Word in Advance." Occasionally a novel would protest its inno- 
cence in several of these at once, in which the author would deny that 
fiction was delinquent, and would suggest that the novel could better 
teach by departing from fact. 

But the author did not come right out and say it. Rather, he implied 
it in so many words. When Nathaniel Hawthorne prefaced The Mar- 
ble Fawn (1860) by saying that "the author proposed to himself 
merely to write a fanciful story, evolving a thoughtful moral," he sum- 
marized far more than his own preface to The House of The Seven 
Gables ( 1851 ) . No fewer than twenty-one of James Fenimore Cooper's 
prefaces make the same backhanded gesture, implying that inventive- 
ness has something to do with the didactic ideal. The tone of Cooper's 
"Forward" to The Crater (1848) sounds as though he had been invited 
to say it: "Whatever may be thought of the authenticity of its inci- 
dents, we hope this book will not be found to be totally without a 

James K. Paulding's prefaces were bolder, just as his novels them- 
selves were more fantastic. In Westward Ho! (1832) he asserts: "The 
great aim of the author has been to combine an important moral, with 
the interest of a series of incidents ... as he knows or imagines exist, or 
have existed." But no matter what kind of novel the reader had in 
front of him, he was bound to feel at home in the preface. Edward 
Bellamy apologized for Looking Backward ( 1 886 ) , saying, "The author 
sought to alleviate the instructive quality of the book, by casting it in 
the form of a romantic narrative." These fainthearted prefaces bear 
some vague relation to the classical concept on which so many Ren- 
aissance and Neo-Classical poets had defended their art the concept 
of virtue as the union of the beautiful and the good but the classical 
"virtue" came out merely as wholesome amusement. 

American novelists were far more willing to refute the charge that 
fiction was a waste of time, for here they were sure of themselves. 
Novels were useful, no doubt about it. They were educational and hu- 

Art and Artifact 19 

manitarian, and hence of practical value. The preface to Charles Brock- 
den Brown's Wieland (1798) points out that the novel instructs, by 
illustrating "some important branches of the moral constitution of 
man"; and Brown appeals the merits of his quasi-science to "physicians 
and to men conversant with the latent springs and occasional perver- 
sions of the human mind." 

Brown's preface to Arthur Mervyn (1799-1800) declares his wish to 
instruct his readers in the medical and political aspects of the historic 
Philadelphia plague, and also to encourage an active interest in social 
conditions and, therefore, a sympathy for the less fortunate. What 
could be more practical? Historical novelists were even more comfort- 
able with this sort of claim. After all, they taught history, and could 
alert their readers to the mistakes of the past. 

The humanitarian responsibility was so clear that even the lack of a 
"cause" could not stay it. When that arch-journeyman Timothy Shay 
Arthur was caught without a specific reform in mind, he nevertheless 
knew his place, and explained that Steps Towards Heaven (1858), a 
book of lay sermons, taught "no special theology. ... It has no aim but 
to assist men to grow better, and thence happier." 

These self-conscious prefaces excuse their novels in a manner sug- 
gesting that the authors had some civic duty to perform. By and large, 
they conveyed what was already taken for granted, and as the nine- 
teenth century wore on the numbers of novels containing prefaces de- 
clined from five out of every six or seven, to one out of every four or 
five. The preface to the American novel had virtually ceased before 
Henry James put it to the use of poetics. 

There is a reasonable suspicion that these messages were not always 
heartfelt. John Neal's preface'to Logan, a Family History (1822) be- 
gins, "I hate prefaces." In any case, by the time Edward Eggleston 
wrote, in his preface to The End of the World (1872), that "Prefaces 
are most unnecessary and useless prependages, since nobody reads 
them," the trend was apparent. The popular novel had outgrown the 
need for the precaution of sanctifying what readers wanted anyway. 

After the iSgo's the apologetic preface got a face-lifting. When pub- 
lishers began to use dust jackets to protect books, they discovered that 
the jackets themselves could carry a message to the readers. The text 

20 The Denatured Novel 

on the inside flap of the paper cover, in fact, was more efficient than 
the preface. Unsigned, because it was written at the publisher's direc- 
tion, it could sanctify the novel without any taint of false modesty on 
the part of the author. It could speak with authority as it lied about 
how the book would improve the reader and about the fun he was 
going to have in the process. But the dust jacket blurb came later. 

Meanwhile, the monotone of the prefaces echoed some remarkable 
similarities among the novels themselves. The resemblance of these 
novels to one another is explained by the fact that a commercially suc- 
cessful book creates a market for others like it. Just as simple as that! 
Every publisher follows the leader, even if the leader turns out a 
book on his own list. Nineteenth-century publishers did not labor to 
point out this fact; the general feeling was that they belonged to a 
"profession" whose dignity was paramount. If a publisher had to make 
money he was expected at least not to talk about it. 

But decorum began to change during the iSgo's, and "legitimate 
profit" became an acceptable subject. In his 1938 Bowker Lecture on 
publishing, Frederick Crofts regarded this tendency to follow the 
leader as an obvious public fact. Publishers are always trying to repro- 
duce the appeal of previously successful books. "An innovation in for- 
mat, in unusual locale, or an original approach or presentation which 
results in a best seller is promptly duplicated." This is so, he said, be- 
cause "a publishing house is not an eleemosynary institution, and is en- 
titled to profit from any definite indication of the public's immediate 

In America during the early years of the nineteenth century, the 
commercial success of native novels depended on their prominently 
duplicating their English predecessors. John Neal spoke out on this 
matter too, in his preface to The Down-Easters (1833). "To judge by 
our novel-writers," he said, "we have cottages and skylarks in our coun- 
try; pheasants and nightingales, first families, a youth of a 'gentle blood' 
and a virtuous peasantry; moss-grown churches, curfews and ivy-man- 
tled towers"; and, he added, hardhearted fathers, runaway lovers and 
cruel stepmothers: "anything and everything, in short which goes to 
the ground-work of a third-rate English or Scotch novel." 

Art and Artifact 21 

Poe elaborately classified this imitative substance in a book review 
he wrote for the Southern Literary Messenger, in May, 1835. He em- 
phasized the "succession of dynasties reigning over the regions of ro- 
mance," each characterized by a popular author (the Radcliffe dynasty, 
the Edgeworth dynasty, and the Scott dynasty) and "each like the 
family of the Caesars, passing from good to bad, and from bad to worse, 
until each has run out." When Poe wrote this, Sir Walter Scott was 
the novelist who had most influenced American readers. 

Scott's novels established a demand for comparable fiction which 
would celebrate the beginnings of the new nation. In this connection 
the novels of Cooper, Kennedy, Simms, Daniel Pierce Thompson and 
James K. Paulding were presented. But Scott's adoring public also 
supported his imitators without discriminating between the original 
and the gross copies. The American booksales of the British mannerist, 
G. P. R. James, for instance, exceeded even Scott's. His five-dozen 
novels, more or less, flooded American bookstores in the i86o's. 

James, of course, wrote more than Scott. But the trick was in writ- 
ing more like Scott than Scott wrote, and just such a mannerism ex- 
plains how the preferences of most American book buyers are satisfied. 
The pathology of this literary cancer was conducted by one of the 
ablest of American critics, E. P. Whipple, in North American Review 
(April 1844), who made it clear that James was no "maker" in the old 
significance of that term. He was a literary mechanic working up the 
same raw material into the same shapes. Whipple had a wealth of meta- 
phors to show the poverty of James' fiction. "For the last ten years, he 
has been repeating his own repetitions, and echoing his own echoes. 
His first novel was a shot that went through the target, and he has ever 
since been assiduously firing through the hole." James could accom- 
plish this precision only by forsaking the individuality of his characters, 
and presenting idealized heroes and heroines and unmitigated villains, 
"walking essays on character." 

James followed the plot around, "catechism in hand," delivering a 
homily wherever needed, Whipple said; and his novels showed it. 
"Characters should be exhibited, not didactically, but dramatically. 
We demand human beings not embodied antitheses, or personified 

2 2 The Denatured Novel 

qualities, thoughts or passions.' 7 Fundamentally, therefore, the imita- 
tion was not at all like the original. "In truth, no two writers have less 
in common, in the essentials of their art, than Scott and James." 

Whipple continued his surgery of James's fiction by asking how 
James could succeed. What did James sell that his customers wanted? 
He sold the image of a world where good will finally triumph, but only 
after its feverish, exciting enslavement to evil. "The world to which we 
are introduced is not a free, common world, where there are chances in 
favor both of vice and virtue; but a fenced park, full of man-traps and 
spring-guns. A sort of iron necessity conducts everything. We do not 
feel ourselves safe until we come to the conclusion. A sort of feverish, 
unhealthy excitement is the feeling we experience as we read." To- 
gether, the happy ending and the premeditated jeopardy denied the 
possibility of real characters, but the characters were not important, 
except for the assurance of their ultimate goal. The real attractions 
were the "man-traps" and the "spring-guns." 

James's novels reconstructed the materials of Scott's novels, but 
James put a greater emphasis on the jeopardy of the hero and heroine 
and designed a more explicit attention to their travails. By this calcu- 
lated exaggeration his novels were his own, not Scott's. They happened 
to be like Scott's, only more so. His was an old trick the mannerist 
borrows some prominent characteristics of a predecessor and reinvolves 
them in a new scheme in such a way as to exaggerate them. But be- 
cause it is a new scheme, the exaggerated characteristics have no 
necessity. And this distortion, many times repeated by many authors, 
explains the process of popular literature. 

What the publishing industry has accomplished is a greater effi- 
ciency in this eclecticism. The result of its efforts is always a cluster of 
new books, reflecting the heat of one that has already made good, like 
a constellation around the sun. The warm qualities of popular fiction 
radiate from elements that never change. Always the novels that have 
made good with American readers have in some way involved the 
family scene, the jeopardy of the toiling heroine, and the strain of re- 
ligiosity. The combination of these elements is what Frank Luther 
Mott, in his Golden Multitudes (1947), has called the Home-And- 
Jesus formula. 

Art and Artifact 2 3 

This formula had a great deal to do with the unprecedented sale of 
novels in the 1850*5. It began when George Palmer Putnam's mother 
told him to publish Susan Warner's novel, The Wide, Wide World 
(1850), a story about the moral travails of an orphan which was so 
notoriously successful that publishers lovingly cultivated all the matri- 
archal novels they could find, and sold hundreds of thousands of the 
books of Mrs. E. D. E. N. Southworth; of Maria Cummins and Mary 
}. Holmes; of Marion Harland, Ann Stephens and Marion Mclntosh. 

There was no coincidence about the gender of all these authors, nor 
in the genetic similarity of their books a fact which Hawthorne noted 
as he railed about the whole situation. "America is now wholly given 
over to a d d mob of scribbling women, and I shall have no chance 
of success while the public taste is occupied with their trash and 
should be ashamed of myself if I did succeed." Hawthorne wrote this 
in a letter to his publisher, William Ticknor, in 1855, while his own 
books languished on booksellers' shelves. "What is the mystery of these 
innumerable editions of the Lamplighter, and other books neither 
better nor worse?" he wanted to know, adding: "worse they could not 
be, and better they need not be, when they sell by the 100,000." 

Hawthorne could have discovered from any one of these books the 
mystery of the others. They dramatized the difficulties which beset 
love, marriage and home; they were about the consequence of tempta- 
tion, with plenty of temptation. They all affirmed that human nature 
is intuitively good, that loyalty and courage will win out, that a girl 
can improve herself in society, and that truth will rise again in the 
happy home. In a chapter entitled "Home Influence" of The Popular 
Book (1950), James Hart suggests the lineage of these novels by de- 
scribing some of their other characteristics, or what Frederick Crofts 
would have called certain prominent duplications: "From Bulwer 
came the drawing-room scene, purified and Americanized; from 
Dickens, the pathetic bride, the orphaned child and the kindly ec- 
centric character; from Charlotte Bronte, a watered version of the Jane 
Eyre-Rochester relation; from Mrs. Gaskell, the sens of quiet rural 

But the real stamina of these novels lay in the religion of Good 
Works which they avowed. Humanitarianism, truly a domesticated 

24 The Denatured Novel 

virtue in this nation, marks another, later cluster of pious novels which 
appeared after the publication of Mrs. Humphry Ward's Robert Els- 
mere (1888). This pastiche recounts a tale of a young minister who 
braces up his flagging religious dogma by entering into social work in 
urban slums. It was promptly pirated in the United States, and vigor- 
ously reviewed in the press and in the pulpit; and the author estimated 
the American sales at a half-million copies six months after the first 
American reprint. 

The newly organized American branch of the Salvation Army gave 
a timely appeal to this version of the problem of what Jesus would do if 
He were here, and the booktrade had a marketable subject. In A Singu- 
lar Life (1894) Elizabeth Stuart Phelps Ward wrote about a man who 
left the Church to live as Jesus had, and to do Good Works. Edward 
Everett Hale's If Jesus Came to Boston (1895) debated the merits of 
humanitarianism with William T. Stead's If Christ Came to Chicago 
(1894). But the novel which reached by far the most people was 
Charles Sheldon's In His Steps (1896) which abridged all the doubts 
and preoccupations of Robert Elsmere in favor of a program for admin- 
istrating humanitarian reform. The simple, practical program of a 
whole church full of Sheldon's characters was irresistible. After its 
serialization in a religious weekly, In His Steps appeared in book 
form. Then, after a defect in its copyright was discovered in 1899, the 
book was reprinted by many presses which sold millions of copies. 

There was no doubt about the appeal of Sheldon's book, because 
there was no doubt in it. By expanding one man's humanitarian desire 
into a public-works program, Sheldon managed to enlarge upon 
Robert Elsmere about as far as possible. Subsequent derivations of 
Mrs. Ward's novel exploited some of its other characteristics. For 
instance, the athletic vigor with which the hero performs his newly 
discovered social obligations became a specialty of Harold Bell 
Wright's novels. 

These clustered novels all served the practical purpose of ethical, 
humanitarian education. They catechized the doctrine of Salvation- 
by- Works, a religious heresy in Europe, but turned by the American 
Puritans into orthodoxy. Within two generations of their arrival in 
the New World, they had domesticated Calvinism, taming the terri- 

Art and Artifact 25 

fying doctrine of predestination with a theology based upon a contract 
between man and God. Thereafter, they made sure, man's behavior 
in this world would count. 

The legalistic Puritans had found plenty of precedent, for the Bible 
is full of covenants to prove their case. But as this Covenant Theology 
became popularized during the eighteenth century, it spread beyond 
the Puritan churches. As more and more people practiced the old 
heresy, Good Works became synonymous with Salvation. All by him- 
self, it seemed, man could avoid the fires of Hell. 

This simplistic belief inspired countless social reforms always in 
the name of Christianity during the nineteenth century. Ultimately, 
in the name of Jesus and In His Steps, it even replaced Christianity. 
And there were other ironies in this fire. From the beginning, the 
American domestication had made Calvinism comfortably secular. 
It was no trouble at all, as Benjamin Franklin proved, to merge the 
problem of Salvation with the achievement of success in mundane 
matters. What a coincidence that the author of Poor Richard? s Al- 
manac, that compendium of piety and profit, also took up the cause 
of Pamela, or, Virtue Rewarded! The attainment of Salvation through 
material success is the particular pursuit of happiness that Americans 
have always been anxious to read about. 

Books that gratify this gross inspiration are known in the publish- 
ing industry as "Blue-Sky" books, and they succeed commercially 
whenever the authors themselves believe in it. The knack in commer- 
cial publishing, of course, is to exploit and to establish if need be 
the coincidence of a writer's personal beliefs with the reigning public 
attitude. And if the publisher believes in it too, so much the better. 
His belief will affect the promotion of the book and probably its con- 

The profit in practical religion, by a publisher who believed in it, 
was superbly demonstrated by the publishing house of Thomas 
Young Crowell. Beginning in the iSyo's for nearly forty years the Cro- 
well lists were studded with inspirational books with clusters of titles 
which prominently duplicated one another. T. Irving Croweirs biog- 
raphy of his father, Thomas Young Crowell (1926), explains how 
some of these clusters happened, 

26 The Denatured Novel 

The elder publisher fashioned at least five successful books from 
the personal vision he shared with Mrs. Sarah Knowles Bolton. In the 
fall of 1874 when Crowell read Mrs. Bol ton's magazine article, "Poor 
Boys Who Became Famous," he invited her to write a book to go with 
this tide, and she composed short biographies of twenty-eight famous 
men. "She skillfully traced the secret of their success, and pointed out 
that without work and will no great things can be achieved." Signs of 
the enormous sales of this book showed up on the publisher's later 
lists, which included Lives of Girls Who Became Famous, Famous 
American Authors, Famous American Statesmen, and Famous Men of 
Science all written by Mrs. Bolton. 

Then Crowell received an unsolicited manuscript entitled "What Is 
Worth While," and the booklet he published from this manuscript 
soon grew into the ' What Is Worth While" series, written by a half- 
dozen authors. One of these, Dr. Orison S. Marden, repeated himself 
for several publishers with Pushing to the Front, or Success under 
Difficulties ( 1894) , Architects of Fate ( 1895 ) and The Secret Achieve- 
ment (1898). 

Here is precedent for the contemporary breeding of those happy- 
solemn tracts which imitate Dale Carnegie and the Reverend Norman 
Vincent Peale. But there is more to this evidence. Crowell had no 
interest in novels, or as his biography puts it, "in ephemeral books" or 
in "belles lettres." "He preferred solid books and was chiefly interested 
in those that would inspire or be useful for reference, so that one 
editor was led to say that he 'never issued a book that one is not better 
for having read/ " The condescension to "belles lettres" and the 
scores of all books that were not practical had, of course, forced writers 
to construct all those prefaces to nineteenth-century novels. The sus- 
picion still lingered. 

CroweH's attitude was no exception. Little, Brown's anniversary 
volunre, One Hundred Years of Publishing: 1837-1937, gravely quotes 
the policy of the first partners of that house: "Their publications in 
general literature have been, for the most part, of a grave, solid and 
substantial character, such as works in theology, history, politics, politi- 
cal economy and biography"; but they had no wish to print novels or to 

Art and Artifact 27 

meddle "with those lighter and more ephemeral publications that 
come with the leaves of spring and go with the leaves of autumn." 

Little, Brown repealed this prohibition after 1890, and such was the 
profit in pietism that when the firm published Henry Sienkiewicz's 
Quo Vadis?, in 1897, it launched the biggest and most intensive pro- 
motional campaign for a novel that Publishers' Weekly could re- 
member. In that same year a PW editorial made one of that journal's 
most memorable understatements about publishing when it remarked: 
"We fear it is becoming a trade." After Little, Brown's Quo Vadis? 
campaign, intensive publicity became a familiar device in bookselling. 
By 1920, when D. Appleton announced to the trade that it was now 
sponsoring the novels of Harold Bell Wright, it had to shout to be 
heard over the general promotion. 

The "Bazaar Spirit," the competitive promotion of literature, is 
the most recent stage in the American domestication of the novel, 
causing a sense of compromise, an astonished awareness of something 
lost by the process of commercializing. In one form or another for the 
past seventy years, the editorials in Publishers' Weekly have repeatedly 
asked the question "Is Bookselling Dead?" And for every one of these 
editorials the Readers Guide to more general magazines lists dozens 
of articles which ask the same question about the novel. These articles 
assess fiction in terms of "cultural values" gained or lost and gen- 
erally weigh so much air. 

There is a certain significance in the repeated doom-calling by 
which these commentators announce their subject: "Will the Novel 
Disappear?" (North American Review, September 1902); "Is the 
Novel Being Superseded?" (Current Literature, September 1906); 
"Does the Present-Day Fiction Make for Immorality?" '(Current 
Literature, April 1907); "Is Present-Day Fiction Quite Ephemeral?" 
(Lippincott's Magazine, March 1909); "The Passing of the Novel," 
(Independent, December 22, 1910); "Are the American Novelists 
Deteriorating?" (Ladies 9 Home Journal, September 1911). 

It was no passing problem, for apparently the plight persists: "The 
Break-up of the Novel," (Yale Review, January 1923); "The Fate of 
the Novel," (Living Age, December 12, 1925); "Are Novels Worth 

28 The Denatured Novel 

Reading?" (Nation, February 13, 1935); and "Is the Novel Done For?'' 
(Harper, December 1942 ). Judging from these funerations, "The 
Slump in American Writing" (American Mercury, February 1940) 
has been continuous. Although most of these essayists think the novel 
will survive (even the few who hope it won't), they are all worrying 
over it. More than that, they share an attitude about the novel which 
makes this whole argument monumentally irrelevant. 

These essayists characteristically assume the novel's responsibility 
to "cultural values," which is what all those nineteenth-century pref- 
aces assumed. The "immorality of the modern novel" to borrow a 
frequent phrase from these articles seems always to be eroding the 
national mind. This, of course, is always presented as an urgent matter, 
even though the novel's degeneracy was apparently a foregone conclu- 
sion 150 years ago. 

In this long, morose debate both the attack and the rebuttal have 
assumed that fiction is some sort of public utility. Certainly, this na- 
tion has domesticated the novel by making just such excuses for it, by 
pondering its usefulness to individual deportment or collective reform. 
And the question of whether the novel (like any other public utility) 
should also serve "commercial interests" has intensified the debate. 

But the novel has always been popular, which is to say commercial, 
because it can always illustrate profusely the drama of moral choice 
and consequence. The legacy of moralism for the past two centuries in 
America has pretty thoroughly obscured the nature of fiction. It has 
obscured the reflexive property of the novel form which really is moral, 
Fiction is a special way of knowing. It records the human individual as 
no other expression can. What happens to it is indeed urgent. 

Whatever has happened or will happen to the novel, after all, is 
mainly relevant to the form itself. What fiction does derives from 
what fiction is. Whether the novel has succumbed to commercialism 
as, of course, it almost always has ought at least to be assessed in 
terms of fiction's intrinsic qualities: on the grounds of why it is, and 
how it is, instead of merely its alleged obligation to public service. 

Chapter II 

How to Denature a Novel 

THE public excuse for the novel is that it is useful. As long as it ap- 
pears to teach something practical it is perfectly all right for one to 
indulge in its entertainment, but the real reason for the novel's popu- 
larity is its gratification of the deep human desire to merge oneself with 
the world. The novel finds the terms of the merger. For some reason, 
however, this profound personal need has never been publicly convinc- 
ing in America. It is easier to get away with the polite excuse. 

The justification of the novel had already been made in the 
eighteenth century, by the time Americans began to import this 
product from England. The works of Aphra Behn and Defoe, of Field- 
ing and Richardson and Smollett, and the responses of British re- 
viewers to these works, had already fashioned the theory of the novel 
as a form of instructive entertainment. It had a purpose, after all. 
But in America it was emphatically necessary to spell out this pur- 
pose, to rationalize entertainment on the grounds of practicality. The 
alternatives were clear from the beginning: A novel either served some 
ethical or educational purpose, or it was wasteful. The prefaces to the 
nineteenth-century novels undertook this face-saving task, but they 
were only symptoms of the assurance the novels themselves were ex- 
pected to provide. 

This either/or reasoning had been long established by the time 
publishers began promoting books in the iSgo's. If a novel wasn't 
"serious" it was "popular" or worse, "commercial." Although "pop- 
ular" and "commercial" meant roughly the same, the American 

30 The Denatured Novel 

prejudice that a novel must serve something outside itself has made 
"commercial" a more comfortably derogatory term. Book promotion 
has merely gone along with these concepts, in that publishers have 
accepted the American contradiction of the inherited theory that en- 
tertainment can be intrinsically instructive. 

The novelists themselves had separated entertainment and instruc- 
tion, using the one to disguise the other, and this divided purpose had 
existed for a long while. James Fenimore Cooper, the first American to 
make a profession of writing novels, had devised a simple expedient, in 
his Leather Stocking Tales, which apparently suited everybody: a 
chapter for the reader, then a chapter for Cooper. The violent action 
would move the characters around for awhile, then they would all sit 
down and have a serious talk. 

Cooper was not the only one with a divided purpose. Neal and 
Simms, even Charles Brockden Brown, were all trying to adapt an in- 
herited form of entertainment to their own sober celebrations of the 
new American republic and the possibilities it held for the individual. 
As often as not, they disguised their philosophical indulgences within 
stories which had nothing to do with their opinions, using the form in 
spite of itself, as Melville later did in Pierre (1852), and as Hawthorne 
tried to do in The Scarlet Letter (1850) with his disarming essay 
about the Customs House. By the end of the century, with the notable 
exception of William Dean Howells, the authors who wrote about 
fiction had firmly polarized the concepts of "serious" and "commercial" 

Frank Norris repeatedly distinguished between literary fiction and 
popular fiction. Literary fiction he equated with art, which he called 
"serious" and "honest"; popular fiction he called "copyism," and said it 
was written by timid businessmen. He insisted that art and commodity 
excluded one another. Hamlin Garland thought the same. The dis- 
tinction between "serious" and "commercial" literature, in fact, con- 
tinues unchanged from the commentaries of Garland and Norris 
through Theodore Dreiser and F. Scott Fitzgerald to James T. Farrell 
and John Marquand. Just as the earlier nineteenth-century apologists 
had found "serious" fiction ethically useful, so these later novelists 
have pronounced "popular" fiction to be merely a commercial device. 

How to Denature a Novel 31 

The trouble is that each of these terms alleges several things at once 
about the intention of authors and the tastes of readers, but the terms 
do not really distinguish anything about the fiction iself. One fact is 
certain: a "serious" novel is not a "commercial" novel which did not 
sell very well. The novel form is popular and always has been, and 
novels that are published are necessarily commercial. Therefore, the 
term "serious" has some special distinction, but most people have 
never been able to agree on what this distinction is. 

As for distinguishing between the tastes of readers, the terms "seri- 
ous" and "popular" cause more confusion than they settle, particularly 
when they are assigned on the basis of book sales. Sales records in book 
publishing are notoriously individual, inaccurate and incomplete. The 
book trade never has had a record of book sales which matches books 
published. This is just a statistician's dream. Only recently, in fact, 
could the book trade match these two unknowns by subject category. 

Most publishing houses have always been privately owned and not 
required to make their book sales public. Even if they were required 
to (and even if they had the information ), it would now take a mechan- 
ical brain to order the confusion of "limited," "trade," and reprint edi- 
tions. A publisher will announce a book's sales only to his own ad- 
vantage. Even then his public affidavit is a merchandising tool inten- 
tionally approximate, the better to impress you. "Seven Big Printings 
in Just Three Months!" the ad will say; "Now in its Fortieth Thou- 
sand"; "Destined to be One of the Top Best Sellers." The fact that it 
is "destined to be" should make you pause. 

When an editor can say of a book, "It was quite a best seller," 
clearly the term is tired. "Best seller," in fact, has been so misused 
since the Bookman first coined it sixty-five years ago that it more often 
expresses a hope than a fact. Since retailers often list slow stock as 
"best selling" in order to move it off their shelves, the term is a poor 
gauge of popularity. 

In common usage "popular fiction" condescends, implying that the 
public somehow makes a book worse by buying it in quantity. This con- 
descension belittles the best-selling novels of Harold Bell Wright, 
Gene Stratton Porter, or Mickey Spillane. But what about the best- 
selling novels of Harold Frederic, Thornton Wilder, Sinclair Lewis or 

32 The Denatured Novel 

Ernest Hemingway? In the same way the term "serious fiction" usually 
confuses the work itself with its author's reputation. 

Take War and Peace, for instance, which everybody agrees is a 
"serious" novel. If you want to say this in a "literary" way, no one will 
argue. Its world is abundant. It carelessly piles up incidents so as to 
dramatize the whole sweep of history, yet it offers an extraordinary 
sense of pertinent detail. During a dozen years in a nation's life it con- 
siders the significant circumstances of a society falling apart and the 
significant problem of discovering whether an individual can afford 
merely to survive. No one would disagree with these claims for War 
and Peace. But it happens they are all borrowed from reviews of Gone 
with the Wind, a "popular" novel if there ever was one. 

Nobody could ever guess the vast difference between these novels 
by reading book reviews. Many generations have decided that War and 
Peace is one of the great novels of all time, while Gone with the Wind 
has already passed into limbo twenty-five years after its publication. 
This simple fact of public response is a timely reminder of the loose 
language of most book reviews. No one even mentions Gone with the 
Wind any more, except everybody in the book trade. This is another 
part of the problem. The way books are defined in the book trade has 
nothing to do with their literary qualities. The literary scholar and the 
bookman talk shop together about as ineffectively as a botanist and a 
florist discussing flowers. 

Communication between the two is difficult at best, but some im- 
portant concepts require the language of both the scholar and the 
bookman. The denatured novel, for example. In order for anyone to 
talk about the denatured novel it is first necessary for him to be clear 
about the novel and its nature. But in what language? The bookman's 
terms, like the scholar's, are often not intended for public communi- 

Like every other business the book trade has its "inside" words. 
There is no public equivalent, for instance, for the noun "one-shot." 
A "one-shot" is a book that is marketed sometimes even written 
to exploit a timely interest. The promotion of the book, therefore, will 
be based on the decision to print a large first edition. A one-shot is 
journalistic; it depends on news value and opportunism. If two of a 

How to Denature a Novel 33 

man's plays were to make a huge profit on Broadway and cause a lot 
of talk about him shortly after he had died, as in the case of Eugene 
O'Neill, you might expect a lot of one-shots about him. 

If this celebrity were not a playwright but a general or an athlete or a 
politician, the shot would be louder. But this is a relatively precise term. 
I once heard a salesman refer to a meretricious novel as "a piece of 
pork"; an editor at another house called it "sheer hamburger." It is 
somehow symptomatic of the scholar's semantic problem that in Pub- 
lishers' Row "pork" and "hamburger" are the same. 

The trade has no official vocabulary. It has had instead, since the 
1890'$, a public parlance and a great many private ones. An editor uses 
one vocabulary when he talks to an author about a book and quite an- 
other when he presents that book to his sales department. They are 
two versions of the public vocabulary the marketese with which he 
sells his books. This is a language of classifications which implies merit 
without having to specify any reasons for the merit. 

The terms "serious" and "popular" belong to the trade's public par- 
lance, but they denote no formulas. Nor do they suggest any judg- 
ment of performance. If they have turned up favorably in book re- 
views, the publisher will always pass them along in his ads, but to him 
"serious" and "popular" are merely classifications of a market. In book- 
selling they are accessories after the artifact. Their familiarity makes 
them useful in describing both the novel and its denaturing. Anyway, 
since commercial publishing is responsible for denaturing the novel, 
its own marketing terms might better describe the process. 

The book trade's functional view of fiction is that fiction is merchan- 
dise. Another functional view, which the book trade inherited and has 
had to accommodate, is that fiction must be practical; in America this 
means that fiction must contain useful information and serve the 
cause of ethics. For different reasons, therefore, the book trade and the 
American public have sold the novel short. They always try to de- 
nature it. 

But the novel, being fiction, need not live down to this condescen- 
sion. On the contrary, no mere functional view of fiction can take it all 
in. Fiction is more than utilitarian. We do not use fiction, we need it. 
It is everyone's continuing preoccupation. Being small and fallible, 

34 The Denatured Novel 

every human being compensates with fiction, inventing a world that 
will acknowledge him according to his own desires. He spends his life 
keeping that world in repair. 

Fiction is the fashioning of an occasion appropriate to one's deepest 
need, which is to justify oneself. It is a kind of knowledge we cannot 
come by in any other way. No one can state his awful dereliction, so 
he curses or prays, or both, and builds a fiction. He realizes a cast of 
characters and a narrative. What his characters are and what they do 
and feel represent all their author did not know he knew. 

Fiction is a bizarre exaggeration of mere actuality. The author must 
exaggerate in order to communicate at all; otherwise his fiction would 
not be necessary or believable. But it still must satisfy one's knowl- 
edge of the way things happen or one's feeling that they could not 
have been otherwise, given what one knows or learns about the char- 
acters. Therefore, a kind of necessity or inevitability rules it, in that the 
action follows given conditions, and nothing else could have hap- 

Since these conditions lie partly in the nature of the character, the 
inevitability depends partly on him. Cinderella, for example, has no 
significance apart from the circumstances she helped to produce, or 
their resolution which justifies her nature. Like any actual person, 
therefore, the character in a novel is a circumstantial being. His ex- 
istence is an acute metaphor of the human condition, for unless he 
participates in the affairs which contain him unless he acts the 
fictional character has no life at all. He must have efficacy, or we will 
have none of him. 

The necessity of fiction, therefore, lies in its believable commentary 
on more than itself. In our own lives we cope with things-as-they-are by 
assigning them value according to the way they impinge on us. From 
our involvements we derive our "values": that is, our convictions, our 
dogmas, our ideals which have shown stamina in the course of still 
other involvements. Certain of these ideals with particular stamina 
seem to be absolute values. But since we derive them only from our 
own experience, most of what we call values are fragmentary and are 
likely to be at odds. Our just deserts are often contradicted, and this 
is why we need fiction. 

How to Denature a Novel 35 

Our recognition of conflict in actual experience is itself a reordering 
(an evaluation) of actuality. Through the liable character, the kind 
of knowledge fiction yields is the nature of one's relativity to everything 
else. Just as fiction in general is one kind of knowledge about relativity, 
so the novel in particular is one way of knowing. (A lyric poem or an 
epic or a drama are other ways.) A novel is a long narrative fiction 
which appraises at the same time that it represents the mutual involve- 
ment of characters. 

The essence of the novel lies in the way it represents; in the way it 
repeats and amplifies its substance to appraise it. It repeats and ampli- 
fies a particular metaphor. All fiction represents by metaphor, by anal- 
ogy to things known; and the substance of all narrative fiction is the 
metaphor of conflict. The novel's particularity is its appraisal of char- 
acters by a various repetition of conflict. 

The metaphor of conflict the encounter convenes opposing 
forces (emotions, traits, and values, persons and circumstances, or all 
of these) in such a way as to localize one's feelings. Conflict tries the 
persons and their values by their actions. Cinderella meekly dressing 
her step-sisters' hair dramatizes the first condition of the beatitude 
which that tale represents. 

Action characterizes the heroine of another fiction, The Scarlet 
Letter, quite otherwise but just as surely. Hester Prynne first appears 
leaving the jail defiantly displaying to the townspeople the scarlet 
letter on her bosom. Each of these heroines reveals herself by what she 
does, because the gesture of each responds to her involvement. It 
makes no difference whether the heroine runs away from the dance, 
like Cinderella; or faces the music, like Hester Prynne. The action al- 
ways responds to something. 

The metaphor of conflict reveals and evaluates because it establishes 
the cause of action. This is why we know more about fictional personae 
than about persons. In actuality we only vaguely realize most of the 
relationships involving us. But conflict organizes actual confusion. It 
clarifies cause, creates the necessity of choice, and therefore makes 
action a response. 

Fiction has a capacity for conjecture and discovery about human 
beings beyond the actual and customary extent of one's knowledge. 

36 The Denatured Novel 

Even more, it stresses what it clarifies. The need for resolution inheres 
in any complication, and by emphasizing complication the metaphor 
of conflict accentuates the need to resolve. Thus, in most versions of 
"Cinderella" the palace ball, which poses the girl's irreconcilable con- 
ditions, occurs for three nights. 

But "Cinderella" is only a tale. A novel differs from it by the breadth 
and depth its length allows. With a plurality of characters the novel 
repeats the fact of involvement. It proceeds by addition or by accretion. 
It gathers analogues to its subject so as to represent an emerging con- 
flict by means of many relevant and contiguous encounters. 

The novel's accretion is not just quantitative. Its process can be de- 
scribed by a singular verb: the novel intricates its conflicts, so that one 
reinforces the others. It is a system of repeated conflicts which variously 
echo one another; each new encounter evaluates the others. This is the 
way in which the novel appraises as it represents. 

This developing of involvement is a way of making the most of 
the least material. In painting, for example, a basic fact of composition 
is that the parts repeat one another, not literally but with variations. 
In music, sequential treatment repeats a motif elsewhere on the scale, 
or by another voice or instrument. The most familiar form of repeti- 
tion in orchestral scores is the theme-and-variations, although its most 
intense use is probably in the fugue, which counterpoints its parts. And 
so in fiction: the novel builds a system of comparisons; its conflicts 
test opposing values and also, by analogy, one another. Like parallelism 
in painting or sequential treatment in music, the dynamics of a novel 
exceeds mere restatement for its own sake, but uses it variously to de- 
velop an idea. And appraisal inheres in development. 

This developing involvement occurs in The Scarlet Letter. The 
novel dramatizes sin and confession in a God-centered world. Hester 
Prynne has already borne a child to the Reverend Dimmesdale when 
her husband, Chillingworth, appears on the scene, and the sins which 
the novel dramatizes are committed after the adultery. Dimmesdale 
fears to admit his paternity; Chillingworth vows to punish him by tor- 
ture; and Hester complies with Chillingworth's trespass. 

Each character experiences the crisis of confession in a way that 

How to Denature a Novel 37 

variously repeats the others. Hester's private confession to Dimmes- 
dale of her compliance in -his torture begets Dimmesdale's confession 
to the world. But the shock kills him, and so Chillingworth is denied 
the opportunity to shrive himself. Chillingworth's failure to confess 
most certainly damns him. 

The dialogues which render these encounters (between Hester and 
her child, between husband and lover, between Hester and her hus- 
band and then her lover) gradually reorder the early encounter be- 
tween Hester and her community into a conflict between pride and 
humility in everyone concerned. What gradually emerges is the em- 
pirical proof of the human dereliction. But because this is fiction the 
representation cannot be stated. When Hawthorne tried to state it he 

In his conclusion he selected a moral and exhorted the reader to 
"Be true! Be true! Be true! Show freely to the world, if not your worst, 
yet some trait whereby the worst may be inferred." But however rele- 
vant his message, its statement terminates the possibilities of the fic- 
tion. It scarcely matches, or even connotes, the dynamics of the book. 

Mere accurate statement fails to match the amplitude of any fiction, 
no matter what the fiction's intricacy. This is equally true of a much 
simpler novel, Margaret Mitchell's Gone with the Wind (1936). In 
this book the warring elements of an entire society convene precisely 
in the heroine's struggle to find both love and security within the nar- 
row limits of respectability. The novel contains a truism which seems 
to stitch it all up. One of Scarlett's many counselors, a grande dame 
on a neighboring plantation, tells her, "It's a very bad thing for a 
woman to face the worst that can happen to her, because after she's 
faced the worst she can't ever really fear anything again." But this cal- 
culable wisdom cannot sum up Gone with the Wind any more than 
Hawthorne's exhortation to be true compends The Scarlet Letter. 

There is more to say about these novels than that they are both fic- 
tion, They are superficially similar but profoundly different, and this 
fact reveals a lot about the nature and the denaturing of the novel. 
The similarities are obvious. Each heroine has been cast out of her 
community: one, for adultery; the other, for stealing husbands and for 

38 The Denatured Novel 

breaking most of the lesser rules for a lady; and both are punished. 
Both novels are didactic, as all novels are, but Gone with the Wind, 
the popular novel, is far more doctrinaire. 

Despite Hawthorne's tiresome moralizing in The Scarlet Letter, 
the moral issue of Gone -with the Wind is clearly more explicit. We 
know exactly where the heroine stands at all times and where she 
ought to stand. No fuzzy equivocations mar Scarlett O'Hara's indict- 
ment; her retribution is tidy and immediately satisfying. 

The difference between these novels lies in the moral involvement 
of The Scarlet Letter and the moral posture of Gone with the Wind. 
Hester's violation of Dimmesdale's soul makes her suffer the fears of 
the damned. But Scarlett O'Hara, who has violated many individuals 
in her attempt to get another woman's husband, suffers merely the 
frustration of never getting what she wants. What seems like her just 
deserts in Gone with the Wind is really only an expedient like the 
prosecution of a gangster for income-tax evasion. 

One novel develops an involvement comparable to the unresolved 
complexity of the world. The other novel evades complexity and sim- 
plifies its subject. In The Scarlet Letter Hester's apprehension over 
the nature of her illegitimate child leads her to interpret the child's in- 
cessant questions about Dimmesdale and Chillingworth as God's own 
ministrations. Hester's inference of divine revelation is what finally 
discovers the analogies in all of the conflicts of the novel. But in Gone 
with the Wind there is no such developing involvement. Its heroine's 
conflict is merely but repeatedly asserted. In Gone with the Wind 
there is a much clearer conflict between a heroine's instinctive desires 
and her inhibitions. Her society will not countenance her only means 
of acquiring the man she wants. But the dialogues between the heroine 
and the other characters, which render this conflict, never get any- 
where. Scarlett repeatedly rejects everyone's advice. The trouble is that 
everyone's mind, including hers, is already made up. Nothing changes 
except the scene. 

The simplistic treatment and the moral posture go together in pop- 
ular fiction, and they have nothing to do with the agony of appraisal. 
Nothing ever changes in Gone with the Wind because the heroine's 
ideal of happiness is unobtainable. But the dream of it is durable. As 

Hoiv to Denature a Novel 39 

long as there are new events to sustain it, the dream goes on and on. 
That is why so many millions of people have sat up all night with it, 
indulging in the heroine's miseries, anticipating her success. The goal 
in The Scarlet Letter, however, is not happiness but salvation and the 
awful crisis of self -appraisal. 

Hawthorne's novel is a re-creation of the values of the seventeenth- 
century Puritan community two hundred years after the fact. Ac- 
cording to the doctrine of this community salvation is far more pos- 
sible than happiness on earth. Like the dream of happiness, it is also 
based on human failure, but achieving Grace requires at least a life of 
continuous and agonizing doubt. This takes work of the sort that denies 
happiness one reason why the quest for salvation often degenerates 
into a misguided pursuit of happiness. 

If Hawthorne had really wanted to match the success of that "d d 
mob of scribbling women" he complained of, The Scarlet Letter 
would have needed a thorough revision. He had an enchanting situa- 
tion: his heroine had a husband at home and a lover-at-large (this 
caused the book to be banned in Russia and the film to be bowdlerized 
in the United States, so he was on the right track), but he spoiled it. 
Hester had already had her lover, thereby removing the delicious frus- 
tration on which the popular heroine's dream of happiness thrives. 

What small popular appeal remained after that error Hawthorne 
also ruined: Hester was not at all concerned about the security of her 
home or about becoming acceptable to the community. She was, in 
fact, accepted long before the book was over. Meanwhile, the author 
made it obvious that the real subject of his novel was the awful crises of 
human dereliction. As far as book sales were concerned, Hawthorne 
botched it. 

This is merely another way of saying that Gone with the Wind, the 
popular novel, offers an utterly simplistic view of the human condi- 
tion. It has stylized the metaphor of conflict. It has made the conflict 
tidy, explicit and literally repetitive. This is how to denature the novel: 
attenuate or thin out the system of analogous conflicts. The terms 
"serious" and "popular" fiction bear on this fact. The "popular" novel 
is usually a denatured one. 

Ideally the novel accumulates thought and action in such a manner 

40 The Denatured Novel 

as to exploit analogies to a conflict, and so appraises as it represents. 
Addition of any new analogous substance necessarily moves attention 
from one point to another and forces comparison of some sort. This 
psychological movement is peculiar to each fiction (and different, for 
instance, in The Scarlet Letter and Gone with the Wind) . The simpler, 
stylized comparison always marks the popular novel. 

Narrative is what moves the attention. Bookmen in the trade refer 
to "story," which means that quality resulting from a stress upon the 
narrative. The "story quality" exercises the desire to resolve things. 
It exploits and satisfies a sense of destination: the universal feeling for 
that consummate end toward which events tend to move. With this 
sense everybody transforms his own actual experience. Thus, the 
"story quality" the emphasis on narrative readily distinguishes one 
novel from another. It distinguishes Moll Flanders, for instance, from 
Pamela; from Madame Bovary or Sister Carrie; from Marjorie Morn- 
ingstar; from The Scarlet Letter or Gone with the Wind in fact, 
each of these from the others. 

This is an aesthetic, not a historical, matter. Two novels by one 
writer can differently exploit the narrative. Henry James's The Am- 
bassadors, for example, which appraises Lambert Strether's growing 
realization of certain circumstances, shows almost no concern for re- 
solving the situation in which he finds himself. But James's The Ameri- 
can has a strong sense of story: of the need to improve the unfortunate 
state of Christopher Newman's courtship; so strong, in fact, that 
James's conclusion in terms other than the story's anticipation becomes 
an anti-climax. 

The specious clarity of Gone with the Wind, repeatedly resolving 
the same conflict, is the result of this story quality. The way a writer 
exploits this emphasis on resolution, in fact, controls the scope and 
therefore the meaning of his fiction. Two novels might appear in all 
other respects to be extraordinarily similar, but their differing empha- 
ses on the resolution of the same conflict will make their entire sub- 
jects different. 

Two such novels are The Plutocrat (1927), by Booth Tarkington, 
and Dodsworth (1929), by Sinclair Lewis. They were contemporary 
and both were best sellers. They share the subject of the American 

How to Denature a Novel 41 

abroad; they both endeavor to define the American's qualities, which 
they commonly assess by European standards of behavior. Each novel 
characterizes its American hero Earl J. Tinker in Tarkington's, Sam 
Dodsworth in Lewis' by comparing him with his opposite, and so 
demonstrating assumed national traits in a tabloid fashion. 

These novels even arrive at approximately the same estimation of 
the heroes. Sam Dodsworth, in fact, explicitly admires his predecessor, 
Earl Tinker. All in all, their similarity might at first suggest an extraor- 
dinary coincidence in literary history. Yet Lewis' novel presents a 
symposium of irreconcilable attitudes, and Tarkington's novel resolves 
the debate. 

The difference between the novels is not philosophical but formal. 
Tarkington's novel develops a conflict in terms of Earl J. Tinker's 
journey across North Africa, with each episode disclosing some new 
realization about Tinker by witnesses who come to scoff and stay to 
praise. The money this Rotarian showers on the province of Algeria 
somehow validates his innocent, outlandish opinions about the cul- 
ture of the Roman Empire. Tinker inspects ancient sewer systems and 
reconstructs the Roman culture in terms of Saturday-night shopping, 
bond issues for war monuments, and public utilities. He resembles first 
a careless barbarian, then the Roman himself, and finally a formidable 
priest of civic religion. This contradiction between what Tinker unhap- 
pily seems to be and what he ultimately turns out to be, prods atten- 

Tinker's low estate in the opinion of more pretentious Americans, 
at the start of the novel, obviously needs correction, but Lewis' novel 
omits the discrepancy between what seems and what is. Dodsworth's 
journey from Zenith, U.S.A., to the Old World causes the reassess- 
ment of two cultures, each in the other's terms. But the fact of the 
matter the mutual differences between Americans and Europeans 
prevails because each apologist argues from only a limited understand- 
ing of the other's culture. Dodsworth comes to realize this, but nothing 
needs to be resolved. His final realization is quite consistent with the 
situation in which he started. It merely explains the antipathies which 
have obviously existed throughout Dodsworth's journey. 

Tarkington and Lewis wrote a great deal of apparently similar fie- 

42 The Denatured Novel 

tion, in that both worked with large, obvious subjects: the complacency 
of the small-town mind, the mores of business, and the confusion of 
ethical ideals in America. Tarkington anticipated Lewis in all of these 
ideas, just as The Plutocrat preceded Dodsworth, but they wrote from 
completely different views of their common experience. In his best 
novels Lewis developed these static ideas, with many analogies, into 
resonant fiction. When he tried (as he did once or twice) to pare 
down a novel to a narrative which had to arrive somewhere, he wrote a 
flat, dull novel. Lewis never could write a story like Tarkington's. Nor 
did Tarkington ever win the Nobel Prize. 

The story quality, as in The Plutocrat, conveys the sense of becom- 
ing. It does so by violating one's natural inclination for order, by dis- 
turbing the stability of any given situation. The habit of representing 
disturbance by the metaphor of conflict is a token of our relative view 
of things which impinge on us, which makes fiction necessary in the 
first place. So, in the merely relative view, whenever one's convictions 
are forcefully contradicted they lose their comforting surety until such 
time as they can again prevail. 

The humiliations forced on the virtuous Cinderella are intolerable; 
so are the ignorant opinions which contradict Booth Tarkington's 
Earl Tinker. Something must be done about it. Likewise, the ignominy 
which the townspeople heap on Hester Prynne contradicts her ad- 
mirable self-sufficiency and honest devotion to her child; and the mores 
of Scarlett O'Hara's world suddenly seem to deny her preservation. But 
the only way to relieve the discomfort these given conditions cause is 
to project them into an action which will prove them wrong. One 
needs to complete the story in order to gratify his wishes for the way 
things ought to be. 

To some degree this reaching sense conveys all of fiction. But the 
story quality, exciting the need for stability, is a simplifying force. It 
tends to ignore variation or development, to override the complexity in 
any arrangement of conditions. The degree to which the narrative pre- 
sides over the fiction that contains it is what distinguishes a resonant 
novel from a denatured one. An author can exploit this sense of becom- 
ing merely for its own sake, as in most denatured novels, or he can use 
it as the means of development, the means of intricating any material 

How to Denature a Novel 43 

into some reflexive structure. The sense of becoming can be diverted 
into a resonant system of analogies, which is the essence of the novel 

The comparison between The Plutocrat and Dodsworth is no special 
case. The distinction between a dominant narrative and a resonant 
system of analogies discriminates among all novels that seem otherwise 
to be similar. A well-established category of merchandise in booksell- 
ing, for example, is the religious novel (the historical novel with a 
Biblical subject); and the peculiar nature of the invented substance in 
religious novels clarifies this distinction. 

The New Testament usually furnishes the conflict, which they all 
dramatize, between the Christian way and the human way. Three 
novels about the last teachings and the death of Christ illustrate how 
differently the same historical evidence can be fictionalized. The Naza- 
rene (1939), by Scholem Asch, is a resonant system of analogies whose 
subject is the human dilemma. The Big Fisherman (1949) by Lloyd 
Douglas, and The Silver Chalice (1952) by Thomas Costain are dom- 
inating narratives which solve the dilemma. 

The Nazarene is the most fantastic and the most believable. Its 
action moves back and forth between Israel in the year of the Cruci- 
fixion and a modern European city, whose inhabitants recall the his- 
torical events. The modern characters are reincarnations of a Roman 
soldier and of a Jew who took part in the original drama and is now 
engaged in translating a manuscript of the Gospel according to Judas 
Iscariot. The reincarnation amplifies the terms of the original conflict 
and perpetuates the old dilemma. Hence the novel's illusion of a con- 
tinuum. Nothing is resolved, but everything is realized. 

The Big Fisherman is more cautious than The Nazarene, more au- 
thentic in its details, but it fails the illusion of reality because the dom- 
inating narrative insists on a conclusion. The novel tells of Simon 
Peter's relationship with Jesus, but the narrative which dramatizes 
Peter's dilemma is a love story involving an Arabian princess who 
has become a Christian. The whole book hangs on the problem of 
her winning the reluctant prince on her own terms and, of course, 
she does. 

In the same way a love story dominates Costain's religious novel. The 

44 The Denatured Novel 

Silver Chalice sets out to dramatize the irreconcilability of self and 
Christ through the struggle of the disciples to preserve the cup of the 
Last Supper from the Jewish High Priest. But the invented substance 
takes over this idea. It concerns a blundering young man who cannot 
make up his mind about Christianity, a conflict dramatized by his love 
for two women. Early in the novel this unfortunate person declares his 
love to the wrong woman and then marries the right one without lov- 
ing her, and this intolerable beginning naturally precipitates an end. 
His discovery that he loves his wife (hundreds of pages later) is so final, 
in fact, that it pre-empts the struggle of the disciples with which the 
novel started. 

The difference between The Nazarene and these other two versions 
of the human dilemma is the difference between a resonant novel and 
a dominating narrative, or the difference between a round fiction and 
two flat ones. The flat ones are denatured. All three are competent, but 
the two that feature the narrative are simpler and more limiting. This 
distinction is basic; it goes beyond the novel form. In all of the fine 
arts, in fact, the reaching sense dominates what for one reason or an- 
other is called "popular" expression. 

It happens that the terms "serious" and "popular" art in this formal 
sense convey an even more explicit distinction in other art forms, as 
for instance, the distinction between illustration and painting. The 
illustration promises a destination. It reorganizes the actual subject so 
as to urge attention forward to a particular destination. In advertising 
design the destination is a product or a specific idea. The illustrations 
of fiction, in books and magazines, likewise anticipate the climax of a 

Illustration on magazine covers often suggests a whole story in this 
way. Norman Rockwell's covers for the Saturday Evening Post, for in- 
stance, illustrate the pregnant moment that point in time when some 
significance is about to become obvious to the persons in the scene: the 
view, from above a sand lot, of the home-run ball speeding toward a 
window, just before the players become fugitives; or the precise mo- 
ment when the girl discovers her brother reading her diary. 

The significance" of such scenes is not the predictability but the in- 
evitability of a climax. Rube Goldberg makes fun of this inevitability. 

How to Denature a Novel 45 

When he illustrates his contrivance to water a buttonhole flower, in- 
volving eighteen mechanical operations and the conditioned reflex of 
a small seal, the viewer patiently follows the directions of the diagram 
from steps "A" through "S," or whatever, conditioned by the fact that 
the end will come. His is a kind of shaggy-dog art, which exploits the 
desire for completion and then offers an anti-climax. The humor de- 
pends on the ludicrous discrepancy between effort and accomplish- 
ment, but no one would laugh at this without a prior agreement of the 
end in view. 

A painting, unlike an illustration, exploits the sense of becoming to 
create a different subject. It tells no story; it concerns itself, its parts 
and their mutual relationship. A painting represents space and planes 
of energy. Its subject is the amplification and development or contra- 
diction of a curve, or of light or darkness, or buoyancy, or brilliance or 

The idea of pure painting, or non-illustrative art, is not very old a 
hundred years, more or less; it did not develop until there were public 
galleries to encourage it. But even within this short time the medium 
has become its own emphatic subject. The styles of painting in the 
dozen years which bracketed World War I show the preoccupation. 
Duchamp's notorious "Nude Descending a Staircase," like all the 
others which upset the visitors to the Armory Show of 1913, is really 
about itself. Cubism, Futurism and Orphism were attempts in different 
ways to evolve the same illusion the simultaneity of experiences 
which actually occur during elapsed time. 

By correlating several states of being, the painting itself becomes the 
subject. Picasso's "Girl Before a Mirror," whose mirrored image dis- 
torts an already deformed original, has the movement of a continuum. 
The eye's unending shuttle between distortion and distortion-squared 
exploits the sense of becoming and makes the viewer collaborate in 
realizing a new, resonant subject. With similar intent Balla composed 
in one painting many separate stages in the continuous movement of 
the "Dog on a Leash." The painting itself is the totality of experience; 
this is why in Duchamp's painting it does not matter that the nude 
descending a staircase never gets there. 

The contemporary vogue of abstract expressionism, which denies all 

46 The Denatured Novel 

representation, merely emphasizes the vast difference between illustra- 
tion and painting. A certain painting and a certain illustration may 
derive from a common subject. The illustration renders an awareness 
of that subject; but the painting renders awareness of the medium and 
the means which dramatize it. If the illustration succeeds, the subject 
from which it derives comes alive; if the painting succeeds, the medium 
itself becomes vital. 

The same formal significance also distinguishes among musical com- 
positions. The standard juke-box song, for example, (conventionally 
thirty-two bars divided into four sections, three of which state the same 
melody) is a skeletal sonata form; but it emphasizes the melody, and 
shortens the developmental section to the length of each of the three 
melodic statements. Consciously or not, everyone acknowledges this 
stylization when referring to popular "tunes." The accompanying lyrics 
develop a story congenial to the melody and, like the melody, exploit 
a sense of arrival. 

This stress on melody actually emphasizes the movement of becom- 
ing, for it is the property of any melody any succession of tones to 
move toward stability. The ear responds to the four active tones in the 
octave scale by anticipating their resolution to the nearest inactive 
tone. Consequently, certain pulls and tensions inhere in any melody. 
The melody of a popular tune, which normally complies with the ex- 
pected resolution of active tones, is relatively predictable. 

What is called "serious" music, on the other hand, characteristically 
delays these resolutions, continuing to pull and play on the tensions of 
the active tones. This suggests a preoccupation with the medium itself, 
with the relationships between the tones. The essence of jazz, for in- 
stance, is developmental and harmonic, not melodic. By virtue of each 
instrument playing against the rest, the jazz composition continuously 
refrains from resolution. But the music called "swing" is a stylization of 
jazz. It simplifies the developmental quality and emphasizes the mel- 
ody much as Tin Pan Alley modifies the sonata form. 

As for the novel form, the narrative capitalizes on the need of an end 
to things. Unburdened by interference, it correlates the universal desire 
to correct the present by projecting it into the future. But any develop- 
ment along the way which delays the resolution of the narrative frus- 
trates the reader's sense of futurity. The novelist who wants to explore 

How to Denature a Novel 47 

the nature of the conflict itself rather than its resolution, therefore, is 
at odds with most of his readers. 

This has always been the great contradiction in the form itself. The 
novel builds a system of analogous conflicts. It is a means of forcing the 
reader to a larger awareness of a subject. But to accomplish this it must 
work with a conflict the reader naturally wants resolved. This contra- 
diction explains why James Fenimore Cooper made his novels tell two 
stories at once. It explains why The Plutocrat is more immediately sat- 
isfying than Dodsworth, and why The Scarlet Letter was not so popular 
as Hawthorne could have wished. 

It explains why Gone with The Wind and The Big Fisherman and 
The Silver Chalice are tabloid versions of experience. The fact that 
they are caricatures explains their popularity. They are one kind of 
compromise between the novel form and the commercial auspices of 
the artifact itself. 

The book trade has always encouraged the stylization of the novel, 
the thinning out of analogous conflicts to get on with the story. The 
reasons are all a matter of economic incentive. In the present day these 
reasons are complex, but they all lie in the kind of product various mar- 
keters of fiction (the magazine, the reprinter and the film studio) de- 
mand of the trade publisher. And the publisher has always managed to 
convey his desires to an author. 

Even after it is completed a novel can be stylized for profit. It can be 
edited or abridged or adapted to another medium and popularized in 
the process. The original can be simplified: the narrative made to dom- 
inate and the analogies thinned out, so as to shift the emphasis from 
an experience to its resolution. This is how to denature a novel. 

This is also the process by which most contemporary novels are bred. 
Many times repeated, under the auspices of commercial publication, 
stylizing causes a peculiar continuity of the novel form. What repeat- 
edly happens is that many novels stylize a few that have made a lot of 

Stylizing begins with the economic fact that a commercially success- 
ful book creates a market for others like it. By this fact the economics 
of book publishing continually modifies literary theory. But book pub- 
lishing has its own complexities, which in many strange ways and 
often for the wrong reasons have shaped the contemporary novel. 

Chapter III 

The Clay Feet of Polk 

Any similarity between this publisher and any 
persons living or dead is purely coincidental, 
because everyone knows publishers are not real. 

"ACCORDING to legend, book publishers move in mysterious ways. If 
not clairvoyant, we are at least marvelously intuitive, as we select, de- 
fine, and propagate the word.'* So Lewis Gaynor, president of the pub- 
lishing house of Polk & Franklin, sketched the opening of the forth- 
coming Bowker Lecture on book publishing. Gaynor had decided to 
call his lecture "Book Publishing: The Insular Industry." 

"One reason for this legend," he continued, "is the crazy-quilt nature 
of the book industry, It is a congeries of small, diverse, independent 
entities, so diffuse that the solution of one economic problem causes a 
worse one somewhere else. The book trade proceeds by exceptions. 
Most publishers operate as retailers; manufacturers occasionally be- 
come publishers; the economic importance of retailers is not book- 
selling; and we all spend our advertising allotments to impress our col- 
leagues in the trade. No wonder the myth of the publisher's intuition!" 

This is Gaynor's disenchanted view. A year ago when he accepted 
the invitation to add to that distinguished series of annual lectures, he 
had no idea he would make such a frightening statement in public, 
at least. But the more he has thought about the industry's problems the 

The Clay Feet of Polk 6 Franklin 49 

more clearly he has seen that they are exactly like his own. Except for 
the Polk & Franklin books that are sold by direct-by-mail methods, 
Gaynor has no idea who reads what he publishes or why customers buy 
what they do. Polk & Franklin is, in fact, the "most average" publish- 
ing house in the industry. The story of this firm explains the economic 
attrition of literature. 


Gaynor has cause to think as he does about this jackpot business. 
During the past twelve months P&F has made an enormous profit on a 
Deluxe ("limited") Edition of the "Song of Solomon/ 7 profusely illus- 
trated, handsomely boxed, and retailing for $17.50. But Gaynor lost 
his shirt on the best history of the French and Indian War he had ever 
read. The house has also lost heavily on a fine sensitive novel by a 
sensitive young novelist. Also, an expensive book about elephant guns 
has had to be remaindered sold at a fraction of its cost to a mail-order 

The editor of the novel and the book about elephant guns has, of 
course, been fired. He ought to have known better. He has listened to 
an agent's sweet-talk in the hope that the agent might someday send 
him some really salable manuscripts. But this editor's mistake did not 
explain why practically no one bought the really great book about the 
French and Indian War, which Gaynor had personally promoted. 

In the first draft of his lecture Gaynor sneered at the myth of the 
publisher's intuition, by quoting O. H. Cheney's Economic Survey of 
the Book Industry ( 1931 ) . This definitive study of the American book 
trade begins by asserting that "the industry is organized around causes 
and effects about as related as dice throws," and for this reason publish- 
ers themselves have come to believe in their mythical intuition. 

"The book industry," Gaynor quoted, "is generally believed to have 
no rules, only exceptions. That is why so many actions in the industry 
are not only not planned or controlled, but they are not even moti- 
vated. Many are hardly intuitive, nothing but tropisms." 

50 The Denatured Novel 

On second thought Gaynor crossed this passage out. No need to 
flaunt the fact. The book trade already knew it, and the layman might 
get the wrong idea. After all, book publishing has always been a per- 
sonal venture, as a genealogy of publishing houses in the United States 
would prove. It would show the persistence with which firms have split 
apart like amoebas into new and whole organisms. Although publishers 
may now know statistically more about their operations than they knew 
thirty years ago, most of the old vagaries persist, intensified by steadily 
rising costs. The organization of Polk & Franklin testifies to this. 

This firm acquired its present personality after the crisis which began 
in its trade book department* in 1948 and finally led to a change in 
management. In that year the entire industry produced 9,897 books, 
and Polk & Franklin accounted for one per cent of the business. It was 
only one of about 1,100 publishers reporting to the Bureau of the Cen- 
sus in 1948; but to an industry in which a few publishers issue most of 
the books P&F was important. 

In 1948 a few more than three hundred publishers produced eighty- 
five per cent of the total of new books or editions, and only fifteen pub- 
lishers produced more than one hundred. Polk & Franklin's 106 titles 
placed the house modestly among the leaders. In the past fourteen 
years it has kept pace with the industry's increased production; in 1958 
P&F published 1 34 titles. 

The public image of Polk & Franklin carries dignity. The firm has 
always published distinguished books of history and economics, and 
to these the works of a dozen well-known English and American nov- 
elists were added in the early thirties, when it absorbed the list of an- 
other firm. These additions have attracted other novelists, one of whom 
the trade regards as a candidate for the Nobel Prize. Five Pulitzer 
Prizes have already gone to Polk & Franklin authors. The works of the 
two contemporary poets on its backlist have received academic ap- 
proval; and the National Book Award to one, followed by two honorary 
degrees, has eased the awful strain of a losing venture. 

The backlist, as Gaynor likes to point out, is rich and varied. Many 

* "Trade book" is the publishing industry's name for a book sold primarily to 
a retail book store, or wholesaler, for resale to the public, in contrast to a textbook, 
or a "direct-mail" book that is, one designed to be sold direct to the purchaser by 
mail solicitation. 

The Clay Feet of Polk & Franklin 51 

motives have built it: the desire to make money, to attract more good 
authors, to exploit a timely interest, to publish enduring books, to es- 
tablish a certain image of the house in the trade. Were any of its 
editors to make a public statement, however, he would say that Polk & 
Franklin has always tried to publish the best book in any given cate- 
gory. This statement carries a faint odor of respectability. It also makes 
"best" a relative word, and accommodates those inspirational novels 
which continue to sell year after year without benefit of advertising. 
(P&F's unauthorized edition of Charles Sheldon's In His Steps still 
sells five hundred copies a year.) 

Polk & Franklin is proud of its literary reputation. When the house 
advertises an "unconventional" novel to use one of the trade's ambig- 
uous terms the retailer can expect a serious formal experiment, not a 
prurient story. Although institutional advertising notoriously fails to 
sell books, this company has for years purchased space in the literary 
supplements during the first week in July, to inform people that read- 
ing is worthwhile. Its letters to authors explaining the rejection of a 
manuscript invariably cite that manuscript's literary inadequacy in- 
stead of its poor sales potential. 

Although P&F has made its reputation publishing general books 
through the trade, its trade department's sixty titles a year comprise 
less than half its titles and only a quarter of its volume. After World 
War II the firm established a college text department, and a juvenile 
list within its trade department. The company has always operated a 
mail-order enterprise. Not many persons know the proportionate in- 
come from these departments, for Polk & Franklin, with stock held by 
one family, issues no public statement of profit and loss; but it is no 
secret that P&F's bread-and-butter is its mail-order department. 

Direct-mail publishing, a way of making books to order, is the in- 
dustry's most flagrant example of the synthetic book. The publisher 
assembles a mailing list of persons with known interests; he then manu- 
factures the contents of a book to gratify those interests and sends a 
description of it to the names he has collected. It always sells and the 
expense is minimal. The publisher can even use and often does use 
his mailing list to solicit instructions for the next book. 

The mail-order business at P&F is mainly nourished by the Franklin 

52 The Denatured Novel 

Library of Practical Living. This venture makes it possible for the firm 
to lose a certain amount of money in its trade department and still 
show a corporate profit. The importance of the Franklin Library (there 
has never been a Franklin in the firm ) has always been affirmed by its 
founder, the elder Horace Polk. 

Issued in celebration of the company's twenty-fifth anniversary in 
1922, Folk's autobiography explains how his publishing venture began. 
These personal reminiscences of Folk's own success would be at home 
in the Franklin Library. Having worked his way through college, Polk 
took over the management of his uncle's dress-pattern magazine; and 
through this connection the young man learned the virtues of hard 
work and frugal living (he earned eight dollars a week). He found in 
his work still other unique opportunities. When the magazine's printer 
went bankrupt with an issue on the press, Horace Polk received as 
payment for his duress the unbound sheets for an edition of Poor 
Richard's Almanac. Some of these he bound and tried to sell through 
the subscription list of his magazine, but with small success. 

Polk persisted, however, and made what turned out to be a brilliant 
decision. He enclosed the rest of the sheets in a letter to each of his 
subscribers, asking how she would like to have this book brought up- 
to-date. The return mail was so encouraging that he hired a lay 
preacher to write a manuscript to order. Polk entitled this manuscript 
The Complete Home Treasury of Successful Planning for Health and 
Happiness in Your Life. He cleaned up with it. 

The young publisher then set about acquiring a list. He purchased 
two other books: a home nursing manual, which he titled The Plain 
Mans Pathway to Health, and a cook book with a large mailing list 
already compiled for it. He larded the recipes in Bridget's Food of Life 
with inspirational messages. All three volumes, uniformly bound in 
sky-blue, sold as fast as the expanding mailing list could accommodate. 
Other ventures, like the garden manuals Polk occasionally developed, 
prospered by mail order also. Later Polk established a book club to fur- 
nish his subscribers with similar products at a slight discount auto- 
matically. But through the years the Franklin Library showed the great- 
est stamina. 

Contemporary mail-order titles (devoted to practical philosophy and 

The Clay Feet of Polk 6 Franklin 5 3 

religion, to public manners, and to simplified psychology) sell about 
600,000 copies a year. The archetypes are still around. Five times re- 
vised, now illustrated with quaint woodcuts, the cook book has sold 
3,000,000 copies. The home-nursing book has been restyled as a loose- 
leaf manual with subscription supplements. The Complete Home 
Treasury of Successful Planning for Health and Happiness in Your 
Life is still one of the book club's membership premiums. When the 
young members of the editorial staff scorn the founder's enterprise as 
"Folk's Practical Piety," they are likely to be chastised. 

The trade department came to Polk & Franklin in the early thirties, 
when it absorbed the firm of Worth and Loesser, whose list included a 
popular history of American military victories and a best-selling report 
of prohibition and organized vice both edited by Worth. Then 
Worth retired, and Loesser, a literary man, became the editor of Polk 
& Franklin's newly acquired trade department. Loesser brought with 
him the works of a French novelist and three British writers. He also 
brought a passionate interest in American writing, and largely because 
of his efforts during the next twenty-five years, P&F now has a backlist 
of distinguished fiction. 

The fiction list also followed other personal interests. Since the 
1930'$ Polk & Franklin has published sea stories because of Lewis Gay- 
nor's enthusiasm. This began while he was the company's sales man- 
ager. Gaynor had maintained a house in Maine, made his own lobster 
pots, and cruised off-shore. One day an agent had a hunch and sent to 
Gaynor the manuscript of a novel with a Down-East setting. Gaynor 
immediately became a reader and buyer of fiction. For the past fif- 
teen years historical novels, particularly about New England seaports, 
have been prominent on the list. Gaynor has even ventured once or 
twice with novels about Indians, a subject he hopes will permit him to 
cash in on the new interest in domestic archeology. 

Gaynor sees this development of a list of fiction reinforced by re- 
gional and scientific interests as an example of sound marketing. "Ev- 
eryone knows there are not enough book readers," he says. "I learned 
that years ago from Horace Polk. He taught me what to do about it. 
He made me read Brander Matthews on the fallacy of trying to please 
'the taste of the public' as if there were only one public, with one 

54 The Denatured Novel 

taste." Gaynor's standard speech to retail associations, entitled "The 
Big 'M' in Marketing/' develops this argument. 

Any competent book, he maintains, will find its small and usually 
insignificant public or market: regular readers devoted to a given sub- 
ject, plus a few additional gift purchasers. But the problem in selling 
and building a book is to get beyond this limited market, to find 
many small groups of readers who might one way or another respond 
to it. "Call this congregation of many groups the book's Market with 
a capital 'M'. Every book should be an overlap, to capitalize on its pos- 
sible interest to many markets." He explains that the publisher may 
break even on his initial cost by the sale of the book to its small and 
known market, but the book will prosper only as its sales exploit the 
interest of many marginal groups. 

This philosophy of marketing of learning a book's basic readership, 
then tailoring it for sales beyond that readership accounts for the so- 
called "house book" at P&F, conceived by an editor and jobbed out for 
writing. Three or four house books, primarily non-fiction, are contin- 
ually in process at Polk & Franklin. But whatever reassurance they pro- 
vide in calculating cost and profit, their timely nature usually limits 
their period of sale. The house book is a one-shot. 

At the opposite extreme, unsolicited manuscripts, which offer the 
least assurance, deluge the publisher. Polk & Franklin receives about 
2,500 unsolicited manuscripts a year, virtually all of which it rejects. 
Editorial assistants empty the slush box after each morning's mail, re- 
turning the impossible manuscripts to their authors and forwarding 
the few promising ones to an editor for a second reading. Occasionally 
one gets on the agenda of the weekly editorial meeting, and possibly 
one a year will be published. But Polk & Franklin dares not forego this 
expensive procedure which might sometime yield another Gone with 
the Wind. Most publishable manuscripts come from literary agents, 
from published authors, and from ideas conceived in the firm. Consid- 
ering all their sources of supply, the editors of Polk & Franklin say "no" 
ten or twelve times a day and "yes" about once a week. 

One editor's personal convictions usually determine the firm's deci- 
sion to publish any given manuscript. This conviction must concern 
more than sales. Shortly before his retirement, Horace Polk installed 

The Clay Feet of Polk & Franklin 55 

in every office in the company a framed copy of a statement by one of 
the trade's leading advertising agencies. "People Are More Alike than 
Anybody/ 7 says the copy. "A publisher who prints a book because he 
likes it usually finds that many people agree with him." And it con- 
tinues: "A publisher who doesn't like a book but prints it anyway often 
discovers that most people agree with him." Lewis Gaynor has en- 
couraged this idealism in his own way. Every editor at P&F knows that 
he had better not make more than one mistake a year. 

The editor must temper his enthusiasm for a manuscript by consid- 
ering the sales records of others like it. Gaynor calls it "exploring a 
book's possibilities." Since no editor has all the facts since his own 
recollections include only the past fifty or sixty manuscripts he has 
read he must test his convictions against the devil's advocates whom 
he meets each week at the trade department's editorial conference: 
other editors, and the heads of the production, sales, advertising, pub- 
licity, and business-management departments, who pass judgment on 
manuscripts, on ideas for books, and on books-in-process. 


It was in one of these weekly conferences late in 1948, when Gaynor 
was still sales manager, that all the trouble in the trade department 
started, although no one knew it at the time. According to the minutes 
of this meeting, discussion centered on two books: a war novel and an 
autobiography. The novel came from one of Loesser's authors who sub- 
mitted two chapters and an outline of a war story. It was his second 
novel. The first, although well-reviewed, had sold only 4,700 copies, 
but P&F had recovered its costs by selling the reprint rights for a paper- 
bound edition. Even though the house held an option on this new 
manuscript at the original terms, calling for no advance to the author, 
publishing it would be risky. 

The other manuscript, the autobiography of a country banker, had 
been submitted by an agent. One of the younger editors at Polk & 
Franklin reported: "Its earthy observations and brash episodes of small- 

56 The Denatured Novel 

town living are very contagious, but the manuscript needs thorough 
revision." George Tasker, another editor, had helpfully summed it up: 
"You mean it stinks but it's great." 

After a fruitful discussion of dirty books, the group considered the 
problem of how best to collaborate with the author. One way was to 
return the manuscript to the author for rewriting (for which the agent 
would first demand a contract for eventual publication); another way 
was to purchase the idea and have the book ghost-written; or, some- 
where between, to offer the author a contract that would provide for 
the house to be reimbursed out of the author's royalties for the exten- 
sive editorial revision planned by the firm. 

(Had this particular manuscript first been submitted to George 
Tasker, it would have come before this group as a finished product. 
Through an agent, Tasker has frequently signed an author of an ailing 
manuscript to a contract, whereby Tasker rewrites the book for a share 
of the author's royalties and then presents the manuscript to P&F. 
Since an editor must one way or another provide his publisher with 
marketable books, everyone prospers by such an arrangement. Tasker 
has a knack with house books, or what might be called a pure notion of 
book sales. He has even offered to compile and edit a Treasury of Filthy 
Religious Wood-Cuts to prove that P&F could profit enormously by 
tapping the immense reservoir of illiterate book-buyers.) 

Of fourteen books the group discussed, according to the minutes of 
this meeting, seven were rejected and five were tabled for lack of in- 
formation, although in the case of three of these it was decided that 
"Mr. Gaynor's sales force will get a market check," meaning a positive 
endorsement from a few large booksellers on the decision to publish. 

As for the novel, Loesser was directed to canvass the possibilities of 
a reprint edition and of a movie sale. The minutes continue: "Mr. 
Loesser will also discuss with the author's agent the possibilities of 
sharing first-serial rights." This, of course, the agent would refuse if he 
knew his business, but P&F wanted a talking point for any compromise 
that might occur. They would need every break to recover costs on this 
book. As for the banker's autobiography, its editor was "directed to 
negotiate a contract for publication, which reimburses Polk and Frank- 
lin for editorial work." 

The Clay Feet of Polk & Franklin 57 

The meeting adjourned after the business manager "asked the pro- 
duction department to present at the next conference budgets based on 
estimated sales of the war novel and of the autobiography." This last 
entry would startle anyone but a publisher. Each book would have a 
budget before it had a complete manuscript. This budget was intended 
to influence all editorial changes. 

A book's budget is a way of establishing the number of copies that 
must be sold in order for the publisher's investment to be recovered. 
This "get-out" figure is arrived at by dividing the edition's manufactur- 
ing and advertising costs (expressed in thousands of dollars) by the 
publisher's net receipt per copy (expressed in pennies), which is what 
remains after the direct costs of discounts, distribution, royalties, adver- 
tising and general overhead have been deducted from its sale price. 
Dividing the fixed or heavy costs by the net income per copy yields the 
number of copies which the publisher must sell to break even. 

But the costs in this budget, which determine the "break-even" num- 
ber of copies, are themselves projections of an estimated number of 
copies to be manufactured in the first place. In short, the budget begins 
by assuming part of what it proves. It is merely an educated guess about 
where to write-off the product. This necessarily elliptical logic doubt- 
less occurs before the manufacture and sale of any consumer product, 
but at Polk & Franklin the products are books; in these two instances, 
moreover, books scarcely begun before they must submit to a budget. 

P&F accepted both of these fragmentary manuscripts, gambling on 
an estimated sale of each to its own basic market plus other groups of 
possible purchasers. What the publisher accepted was really only the 
beginning of an idea, the rest of which would grow and take shape ac- 
cording to specifications in the budget. Although each manuscript 
originated with its author, the books themselves originated with the 
publisher who had, in effect, to conceive them in space. Gaynor always 
explained to a new editor that planning a book was like designing a 
house from the roof down. He failed to explain the author's problem 
in completing such a manuscript, which is more like building a house 
from the roof up. 

Nothing seemed to be amiss. The minutes of editorial meetings 
through the spring of 1949 revealed progress on both books. As ex- 

58 The Denatured Novel 

pected, Loesser could get no income from magazine rights to the novel, 
although two paperback publishers Monster Books and Cleavage 
Press had expressed interest. With this possibility of bargaining, P&F 
could probably count on a $15,000 advance against paperback sales, 
half of which would belong to the author. Accordingly, the business 
manager approved a budget based on a break-even sale of 7,500 copies 
of a $4.00 novel. 

A large package would have to help justify this high retail price. After 
the success of The Naked and the Dead the year before, the literary 
fashion in war novels called for length and great detail. Loesser would 
consult with his author about building into the story several episodes 
not presently in the manuscript. One of these, the firm agreed, should 
be the American liberation of a brothel. This would give point to an- 
other new episode, when the hero would return to the scene of said lib- 

Since the book's sale seemed largely unpredictable, the advertising 
to the trade would represent it as a "quality" novel a distinguished 
work which Polk & Franklin was "proud to publish.' 7 Now it was time 
to name the book or rename it. The author's title, Toe Hold in 
Elysium, simply would not do. As Gaynor said, they wanted a quality 
novel, not a precious one. The narrative told of a landing operation in 
the Mediterranean, Loesser explained. George Tasker, who had read 
most of the manuscript, said that if a precise title was wanted, it ought 
to be called Honest, I Was Really There. But the minutes state merely: 
"After some discussion, the meeting decided on Red Beach Assault. 
Mr. Loesser will explain the change in title to the author." 

The other book, the banker's autobiography, also progressed. It 
would be tailored to fit a 35<>page package retailing at $4.50. The 
banker had a success story to tell, and since he also had a way with 
homespun maxims, the edition might possibly find a second sale in the 
Franklin Book Club. On the strength of this the production depart- 
ment requested manufacturers 7 estimates of the cost of producing 
20,000 copies. 

With general agreement the autobiography was titled They Bank 
on Me. The editor penciled a memo to the art department, suggesting 
a folksy book jacket; and from several designs he chose a village scene 

The Clay Feet of Polk 6 Franklin 59 

in the style of a cross-stitch sampler, with the title set in Olde Alma- 
nacke type. By July each book had a production schedule for publica- 
tion in the spring of 1950. Under its editor's direction the autobiogra- 
phy proceeded more smoothly than the novel. With the final manu- 
script ready in September, both its dummy and jacket were prepared 
for the December sales meeting. The manuscript of the novel arrived 
late in November. By this time the designer and the copy editor had 
already prepared its jacket. 

Production of dust jackets signaled the start of the advertising cam- 
paigns for both books. Each book's budget provided for an advertising 
appropriation based on an estimate of its sales to retail booksellers and 
to jobbers in advance of publication as low an appropriation as fea- 
sible, considering the estimated life of the book. Polk & Franklin nor- 
mally spends ten per cent of the gross receipts from advance sales on 
advertising the list as a whole, unequally apportioned to individual 
titles. Each book must earn its basic share of this kitty: ten per cent of 
the net receipts of its own advance sale. A percentage of the difference 
between net and gross receipts builds a reserve fund to which the house 
can charge extra advertising as needed. 

Accordingly, They Bank on Me and Red Beach Assault each had a 
basic allotment. The autobiography, with an advance sale of 15,000 
copies, at a net of $2.70 per book yielded $4,050. The novel's advance 
sale of 5,000 copies yielded an advertising budget of only $1,200. But 
gambling on Loesser's enthusiasm (an expensive luxury, as it turned 
out) P&F increased its appropriation by another $1,200 from the re- 
serve fund. Following publication, each title would have to earn its 
advertising by current sales. 

These advertising appropriations at Polk & Franklin proceeded from 
two principles of book selling. One principle is that a book sells pri- 
marily by its own contagion, for which advertising offers no substitute. 
One reader tells another, a process which may take months or years for 
the book to find its market. Advertising can speed this process, once it 
begins, but consumer advertising must follow book sales. Advertising 
can anticipate the book only by announcing it, and then only to the 

The purpose of this trade advertising is to impress on everyone con- 

60 The Denatured Novel 

nected with the book, from author to retailer, how much the publisher 
believes in his product. He also promises that everybody can make a 
lot of money on it. Ads in Publishers' Weekly are part of this campaign. 
But P&F also has a carefully updated mailing list of key persons in the 
trade who receive advance copies of the book. These persons also re- 
ceive a personal letter from the president of the company, which is to 
say one of seven different form letters written in the sales department, 
expressing the president's profound faith in this splendid new book. If 
bookmen believe all this enough to start talking about the book, the 
publisher has spent his money well. 

At Polk & Franklin they call the second principle of bookselling "the 
infallibility of the advance sale": an overstatement of the publisher's 
belief in the reliability of the pre-publication sale to jobbers and retail- 
ers as an indication of its ultimate market. Books' budgets are esti- 
mates; but even an estimate must begin somewhere. More important, 
this advance "sale" influences subsequent editorial decisions. In view 
of a book's sales record a year later, its advance sale will help estimate 
the possibilities of any similar manuscript in the future. 

The advance sale is at least calculable or was, until ten years ago, 
when retailers demanded the privilege of returning unsold books, by 
a system similar to magazine distribution. Considering that a "sale" is 
not necessarily a sale, the degree to which publishers rely on the con- 
signment of their products to establish their costs would frighten most 
manufacturers. In fact, Gaynor has lately given up trying to play this 
guessing game with retailers, and is now experimenting with the 
scheme of sending them assortments of books, like variety cereal pack- 

The advance sale has traditionally become important because of sea- 
sonal marketing and crowding of books. Every P&F title released 
through the trade appears on one of two lists a year. The fall list, aimed 
at Christmas buying, begins with publishing dates in August and with 
books already presented to jobbers and retailers in June. The spring list, 
with publishing dates beginning in February, has already been pre- 
sented to the trade in January. At Polk & Franklin semi-annual sales 
meetings precede these July and January deadlines. 

These sales meetings represent both cause and symptom of Polk & 

The Clay Feet of Polk 6 Franklin 61 

Franklin's frantic seasonal preparations for merchandising books. On 
these occasions the house calls in its travelers who sell to the retail 
trade, for a two-day conference to acquaint them with the new list. 
Polk & Franklin's travelers have always formed the infantry in its sales 
campaigns. With their advance sales to retailers they consolidate the 
publisher's trade advertising. But since their jobs largely depend on 
their customers' confidence, the salesmen also represent the retailers 
to the publisher. The traveler must be shrewd enough to know with- 
out benefit of reading the nature of a given book, why it has been 
published, and how the publisher plans to sell it. 

The presentation aimed at answering these questions comes from the 
editor. From the moment of the book's conception, it has been his job 
to define its qualities and to translate them into selling terms. This 
representation of a book usually requires a further translation into sales 
quotas by P&F's sales manager. With thirty new books at every sales 
meeting, time and attention are limited. Since an average presentation 
lasts only eight minutes, the editor needs some shorthand method to 
reduce the book to a negotiable status, so he expediently classifies it. 

Classifying a new book according to its subject or its theme offers 
the advantage of making its appeal measurable by the sales of past 
books presumed to be more or less like it. But this is the only advantage. 
To the degree that it succeeds in that compromise, classification moves 
everyone's attention away from the book's individuality, away from 
that quality which makes the book vital. Categorizing books contra- 
dicts the industry's maxim that every book is an exception. An editor at 
Polk & Franklin may spend months nurturing a book's peculiarity, 
some precious eccentricity of its author, but at the sales meeting the 
only way he can express its freshness and its difference is by homologiz- 
ing it. 

Tom Loesser always used to chafe at this need for classification. Any 
worthwhile book, he would say, projects a personality, and should not 
be tolled off as merely one of a category. While still a young man, 
Loesser was privileged to know the late Maxwell Perkins and to share 
some of the fascination and the pain of that editor's celebrated mis- 
sion. Perkins had received an amorphous, much-rejected manuscript 
fortuitously entitled "O Lost," and had made its author discipline it 

62 The Denatured Novel 

into Look Homeward, Angel. In the process of making his emotions 
articulate in some more manageable form, Maxwell Perkins had pro- 
jected the singularity of Thomas Wolfe. 

Loesser always believed in the lesson he had learned from Max Per- 
kins. The editor must represent the book at every stage of its produc- 
tion: consultation, copy editing, manufacture, jacket design, advertis- 
ing and selling. Yet, in the last analysis, he must amplify the author's 
intentions in the most commercially feasible way. This contradiction 
always disturbed Loesser. "Novelists don't write categories/' he would 
say. "The question is not what is a pot boiler but who is a pot boiler. A 
man writes as he is. If he has the soul of a pot boiler, he will write pot 

When Loesser spoke like this he spoke like Henry James. But James 
never sold many books. Anyway, book promotion has always worked 
the other way around. A publisher will emphasize a book's similarity to 
others so that he can then show how it exceeds these others in the 
qua^ties they presumably share. Sooner or later the salesman must sell 
the book, having judged it on his own terms. His terms are the sales 
of previous books like it. 

Loesser's notes for Polk & Franklin's sales meeting in the winter of 
1950 yielded 0^11 this point. He later said they had made him retch, and 
Gaynor haci^agreed. After a typed synopsis of Red Beach Assault, 
Loesser penciled these notes: 

1. As you can see, novel concerns Italian campaign in World War II. 

2. Has a love story, like A Farewell to Arms, only it takes place in 

3. Soldier-hero is not running away. It is more positive. He goes off 
to the enemy like the last episode of The Gallery. 

4. A landing at Anzio and suspense of a limited, bitter campaign 
like A Walk in the Sun. 

5. Soldier has no idea where he l*ts into the sense of what is going 
on around him, like The Red Badge of Courage. 

6. Promotion : we will give this shock advertising, like The Naked 
and the Dead. 

7. But our pitch is that it is first and foremost a quality novel. 

8. Will be a paperback edition. 

The Clay Feet of Polk & Franklin 63 

9. Movie rights have already been sold, possible tie-in with Holly- 
wood's advertising. We think can get a Life article on World War 
II films and how this one will be different. 

Shot-gun salesmanship also marked the presentation of They Bank 
on Me. Gaynor remembered thinking that this young editor had 
learned fast. 

You remember impact of Country Lawyer and Country Doctor. 
WARM books!! This banker has watched over his town in the same 
way, and taken care of its people for fifty years. 

This book will break down image of a bank as a big, impersonal in- 
stitution. Banker has his own personal problems, like Charles Gray in 
Point of No Return. 

But in this new book, appeal of the man in the bank is really more 
like David Harum. A kind but shrewd honesty. WARM-HEARTED!! 

National Usury Association is behind us on this one. Its journal for 
July will lead off with a review of the book. And we will advertise in 
Wall Street Journal. 

One other important tie-in: the Southern crisis. This bank is in a 
Southern town and the big episode (synopsis) is about the banker try- 
ing to help the school principal get money for a new building for an 
integrated school. 

We plan to advertise this book as controversial in the South. This 
will go big in East, West, and Northern markets. 

The rape and the lynching speak for themselves. 

Author is already signed up for a series of appearances on college 
campuses. The liberal angle. Publicity Department is already working 
on the endorsements. 

So the books went to market. Both justified the original estimates, 
but with unexpected consequences. Considered an investment in its 
author, Red Beach Assault performed as intended. Polk & Franklin 
sold out the original edition within a year, although scarcely recovering 
costs because of the increased advertising budget. The advance sale to 
the trade, 4,300 copies, proved slightly less than the estimate. Never- 
theless, P&F followed up a half-page ad in Publishers' Weekly with a 
series of smaller ads before and after publication in the New York 
Times and Herald Tribune book reviews and in the more literate maga- 

64 The Denatured Novel 

zines with a small but national distribution. It was also listed in several 
of P&F's other ads in these same publications. 

When the paperbound edition of the novel, retitled Beachhead in 
Hell, appeared a year after publication, the sales of the hard-cover edi- 
tion stood at 7,300 copies. Its trade sales stopped at that point, but the 
paperback went well. Monster Books sold 200,000 copies in the first 
printing; another 650,000 copies in 1951 tied in with the movie. But 
no one had made any money except the author. By June of 1952 the 
reprint sales had barely repaid Monster Books for its $12,000 advance 
against royalties. All the milkmaids had done their job, but there was 
not enough milk to go around. 

Well reviewed and carefully published, the war novel was a succes 
d'estime, a fact which curiously nullified P&F's investment. All had 
gone according to plan, but the author did not see it this way. He 
attributed the difference between the favorable reviews and the small 
sales to the lack of advertising. P&F's capital risk and its negotiations 
with the subsidiary marketers which had enabled the publication in the 
first place made no impression on him. Then, on the strength of this 
novel another publisher offered the author a ridiculously large advance 
for a volume of his short stories, and he immediately challenged P&F 
with the offer. This was an outright trespass. Everyone knew it, includ- 
ing the agent who had promoted it. 

Tasker reminded everyone that this was exactly the sort of raid P&F 
pulled off whenever it could. Nevertheless, Polk & Franklin was on the 
spot. The other publisher obviously planned to steal the author by pub- 
lishing the short stories at a loss. Expensive luxuries, short-story vol- 
umes yield only small sales (2,000 copies in this case would be good) . 
Having published this author's works with as yet no profit, Loesser 
urged him to continue work on his third novel, even offered him a 
$1,000 advance; but P&F "regretfully" turned down the short stories. 
The refusal to lose more money finally cost the house the author in 
whom it had invested. 

They Bank on Me brought surprises too. Advance sales had totaled 
10,000 copies, slightly higher than expected, Fortunately a trucker's 
strike had prevented P&F from supplying the printer with enough 
paper for the first run, and so a second printing before publication had 

The Clay Feet of Polk 6 Franklin 65 

been necessary. Then came the break which made the book. A large 
manufacturing company ordered 7,000 copies at a jobber's discount 
to be delivered in time for distribution as a Christmas bonus for its 
employees. This company's president, like the banker a self-made man, 
had noted the strong preachments in the book about the beauties of 
living happily with less money, and desired to bestow this comfort on 
his employees. 

Timing was perfect. Polk & Franklin printed the copies and immedi- 
ately took full-page newspaper ads to announce the third printing 
within a week after publication. The book, in fact, did sell the 7,200 
copies within the first month; and, still on the strength of its Christ- 
mas-bonus order, Polk & Franklin gambled on another printing ahead 
of sales and another series of ads proudly announcing "Four Big Print- 
ings in Five Weeks/' The lure worked. The book began to sell. Enough 
retailers who had overstocked it reported the book a best seller to give 
it a place on two best-seller lists. Cleavage Press offered P&F $25,000 
for reprint rights. The book sold steadily in the trade edition, and with 
the help of Christmas buying P&F got rid of 35,000 copies. The re- 
maining books became selections in the Franklin Library. 


Polk & Franklin published these two books at a time of rapid changes 
in book markets, when it was becoming evident that trade publishers 
had no business publishing fiction. For one thing, over-production had 
returned with the peace-time economy. During World War II, when 
the rationing of metal and paper had severely limited the manufacture 
of books, publishers had enjoyed a seller's market; any book would sell. 
During the war years the industry had produced an annual average of 
7,800 trade titles. In 1947 the total jumped to nearly 9,200, then to 
1 1,000 by 1950. But customers for these books had failed to materialize. 

A more specific change had drastically affected the sale of fiction. 
This was the tremendous increase in the number of inexpensive paper- 

66 The Denatured Novel 

bound reprints. The adjusted Census of Manufacturers indicated a 
sale of 95,500,000 paperbacks in 1947, sightly less than a fifth of all 
books published in the United States. By 1952 sales had risen to 270,- 
000,000 copies, more than a third of the annual total production. And 
between these years fiction comprised nearly ninety per cent of the 
paperback titles. Clearly, more book buyers wanted their novels in 

Simultaneously during the early post-war years manufacturing con- 
ditions all but prohibited the profitable publication of novels in trade 
editions. Before the War, as late as 1941, a trade publisher could break 
even on the sale of 2,500 copies of a novel, and since fiction had always 
been his most speculative product he could afford to publish many 
titles each season in hopes of one sales winner. But the scarcity of paper 
persisted for several years after the war. Book publishers could not 
compete with the magazines, which were able to buy paper in large 
quantities, or with newspaper syndicates able to purchase mills and 
contract for their entire output in pulp. 

The rising cost of labor also worked against small editions. Make- 
ready time on short printing runs became too expensive, so book manu- 
facturers gave priority to large editions. Delays were costly to the pub- 
lishers. In the early post-war years all costs had increased at Polk & 
Franklin, but the firm clamped down where it could on the cost of 
advertising and the cost of royalties. As a consequence of these econo- 
mies, in the case of the war novel at least, Polk & Franklin lost an 

Rising costs and changing readership had determined most of the 
editorial decisions at P&F in the publication of these two books; and 
their respective sales helped cause a palace revolution in the firm. As 
an investment the Red Beach Assault had failed; the departure of its 
author nullified much of the prestige its publication had brought to 
the house. But, as a one-shot, the autobiography had succeeded. With 
better luck than the novel and with shrewd merchandising this piece 
of editorial carpentry had paid off, earning the house almost $85,000 
from trade sales and another $12,500 in reprint royalties. Because all 
books tend to attract others like them to a publisher's list, P&F could 
expect more of each kind. Some policy had to be set. 

The Clay Feet of Polk 6 Franklin 67 

Everyone at Polk & Franklin recognized these books as belonging to 
different breeds, but neither book was an extremity of its kind. The 
novel was promising but not first-rate. The autobiography was mer- 
chandise but not worthless. Tom Loesser justified publishing both. 
"We want a varied list," he said, "the best of all kinds of books." He 
argued that this was the only sound compromise a publisher interested 
in quality could make. Since a book's sales did not correspond to the 
book's quality in this imperfect world, then competent merchandise 
would have to pay the way for good literature. 

Some of the facts have justified Loesser's argument. For better or 
worse, a publisher's bookkeeping characteristically charges all the cost 
of a book against the trade sales of its first edition, making it look like 
a poor investment. Any income the book earns from a magazine, a re- 
print, a book-club selection or a film is entered in a separate account for 
"subsidiary" income. The income in this second account the publisher 
considers as profit, overlooking the fact that his original investment 
(his trade edition) helped earn this subsidiary income and therefore 
merits credit for a share of it. 

This arbitrary partition between ledger pages often determines edi- 
torial decisions. According to Loesser's thinking, one book had to sup- 
port another. This was the way the publisher could justify "the worth- 
while book of limited appeal." But Gaynor saw it differently. The only 
way to justify a loss was to blame the editor, and axe him. The guillo- 
tine always works. 

Loesser had patiently tried to argue his compromise by analogy. "The 
auto manufacturer doesn't expect the same dollar-return from the 
carbureior and the clutch plate and the hub cap. He sells a composite. 
Each of the parts costs him more or less than average, yet the whole 
product needs all its parts." 

For awhile Gaynor argued with him. "According to your analogy, the 
manufacturer could afford a better clutch plate by making a cheaper 
carburetor. But you don't want to admit it, because if you did, then 
you couldn't rationalize publishing junk." 

The times favored Gaynor. P&F steadily lost money after 1948, and 
the sons and daughters and the new in-laws of the Polk family 
finally persuaded the president to step aside. Upon the retirement of 

68 The Denatured Novel 

Horace Polk, Jr., to a newly created chairmanship of the board, in 
January 1950, Lewis Gaynor was appointed president of Polk & Frank- 
lin. Within a month the business manager became vice-president, and 
the advertising manager took over the sales department. Two of the 
five trade editors were fired, and replaced by younger men from the 
production department. Loesser was persuaded to head Polk & Frank- 
lin's new textbook department; and his successor, with a title of execu- 
tive editor, was hired from Monster Books. 

The consequences of this palace revolution presently appeared. A 
series of memoranda from the president to all department heads firmly 
defined a new editorial policy for the trade department. "We believe 
in making substantial profits on books of which we can be proud," the 
first memo began, "but we are not content with marginal profits/' 
Simply stated, the intent staggered Loesser: every book must make a 
substantial profit. "To guide your thinking along these lines," the 
memo continued, "we will aim at a total minimum sale of $25,000 net 
for each title over a reasonable period of time." The memo suggested 
three years as "reasonable." 

This was sound policy. A book's pricing formula would depend not 
on the number of sales needed to break even but on a stipulated rev- 
enue to be earned by any means whatever. A book priced at $3.50 and 
returning $2.00 net to the publisher would have to sell 12,500 copies 
in order to meet its obligation to the firm; or for every thousand sales 
less than its quota it would have to earn for the publisher $2,000 of 
subsidiary income. This policy would eliminate the marginal book. Al- 
though the president did not bother to explain it, this policy would 
censor future lists at Polk & Franklin before they could even be as- 
sembled. This censorship would take its heaviest toll among novels. 

The effective book selling period of novels (from three to five 
months) is shorter than for other books. Novels would, therefore, have 
less time in the market to achieve the stipulated minimum sale. Unless 
its author's name had sales appeal, or unless it could provide enough 
subsidiary income 'to make up the difference between its actual sales 
and its quota, a novel henceforth would have small chance of accept- 
ance at Polk & Franklin. 

The Clay Feet of Polk & Franklin 69 

Subsequent memoranda elaborated this policy. One of them, en- 
titled "Our Biggest Liability," explained why the expense of marginal 
books exceeded even their estimated costs. "The book with the limited 
market will usually be over-printed to meet our pricing formula, over- 
advertised because it is over-printed, and over-sold in terms of its true 
market. Unsold copies returned to us for credit cause a loss in every 
department that must touch the book." Another memo, entitled "Con- 
servative Terms," informed the editors that advances and high royal- 
ties to authors were no longer automatic. "Polk & Franklin," it said, 
"is not a charitable institution." 

In view of these changes, Thomas Loesser's final transaction in the 
trade department caused him unusual distress. Before the revolution 
he had received the manuscript of a Western novel. It was about a 
rancher but offbeat, what the trade would call a "serious" cowboy story. 
It was a character-study of a man who turned out to be his own worst 
enemy. Loesser had originally liked the outline and had invited the 
author to sublnit more copy. Now, the first five chapters and a precis of 
the rest looked promising. The story began conventionally with a stam- 
pede and a narrow escape and even ended in a chase, but with a differ- 
ence. It all occurred in the man's mind; and Loesser recognized the 
quality by which the trade would label this novel as "serious." The 
story concerned less the events than the perception of these events; it 
was cerebral. 

It was a splendid book and worth publishing, but Loesser knew it 
would never make its quota under the new system without the aid of 
subsidiary income. Since its marketing category disqualified it as a 
Western no reprinter would even consider it until he could see a com- 
pleted manuscript. Meanwhile, the author had run out of money and 
needed an advance in order to finish the book. 

But the change in command intervened. When the new executive 
editor was briefed on these data, he did not need to read the manu- 
script. He dictated a letter of regret to the author, censuring "the 
novel's aesthetic failing." He faintly applauded "the descriptive qual- 
ity," but pointed out that "in the last analysis the characters are not 
quite believable in this context." 

jo The Denatured Novel 


Loesser was not the only lonely man at Polk & Franklin. Its new 
president had distresses of his own. Lewis Gaynor had put his house in 
order only by exchanging compromises, but he felt he had to curb 
over-production too many copies of too many titles which (except 
for the war years) had plagued the firm as long as he could remember. 
Horace Folk's impeccable logic had taught him the consequences of 
over-production; and he had reduced the arithmetic of publishing to 
what he called Folk's Law: "Thou shalt not oversell." 

Gaynor still recalls those two books Polk & Franklin published 
while he was sales manager: the war novel and the autobiography. 
Probably neither one should have been published, although for differ- 
ent reasons. The house had published one for its literary merit and the 
other for profit, but neither had profited Polk & Franklin's list. The 
novel had cost the house an author, and the autobiography had cer- 
tainly not attracted any worthwhile manuscripts. 

Gaynor recalls these events as he prepares his lecture on the sad state 
of the publishing industry. One trouble, he has decided, is that most 
publishers know so little about their own business that they have to 
pretend to be intuitive. For years the trade has ignored the statistics of 
its own operations. Nor can a publisher readily find out what he needs 
to know. The trade's partial and private bookkeeping denies the discov- 
ery of even fundamental statistical knowledge. 

Separately recorded as they are, even the two most elementary facts 
about the industry the number of titles annually published and the 
number of copies sold are mutually irrelevant. The annual inventory 
of new titles in Publishers' Weekly and the periodic tabulation of 
sales in the United States Census of Manufactures classify informa- 
tion according to different subject headings. The integration of text- 
book, trade and mail-order publishing within a single publishing house 
has compounded this problem of finding relevant statistics. 

The Clay Feet of Polk 6 Franklin 71 

These are only symptoms of the real problem, Gaynor concedes 
which is the trade's ignorance of book readers in this country. The pub- 
lisher might make an educated guess about the number or geographical 
location of book buyers for any given title. But about readers, he knows 
almost nothing. For years he has ducked the responsibility of finding 

Gaynor lays no claim to scholarship, but he does have a historical 
sense; and his marginal notes in Cheney's Survey of the Book Industry, 
in a chapter called ''Readers or Book Buyers," indicate that he has 
found the problem. "We are beginning to realize that the processes of 
the book industry do not form a straight line from author to reader," 
Cheney states. "They form a sort of continuous circle within a circle in 
which the reader does (or rather, should) influence every branch of the 
industry, including even the author and the critic." 

The industry has been so concerned with the book that it has forgotten 
the reader. But the influence of the reader on the book must sooner or 
later be understood by the publisher and bookseller and the result 
must inevitably affect the nature of books published, their format and 

their price, their theme and style and, it is hoped, merchandising 


Gaynor has underlined it. "The industry has been so concerned with 
the book that it has forgotten the reader!' 

The inevitable guesswork by which publishers build their lists 
Cheney facetiously calls the "spawning theory of publishing," in a 
chapter called "The Book, the Buyer and the Critic." The whole busi- 
ness is based on guesswork. "The good publisher is the one who guesses 
wrong least." 

Logically, therefore, it appears to be the duty of the publisher to guess 
as often as possible the longer the lists, the more the good guesses. 
The spawning theory of publishing cannot, if it carries its logic to a 
conclusion, reject any manuscript which contains even traces of merit 
or theoretical sales possibilities it might turn out to be a good seller. 

j2 The Denatured Novel 

Cheney explains the relationship between too many books and their 
short life in the market place. The excessive lists of titles seasonably 
issued mean that each book gets inadequate promotion and selling 
time. "Books are cannibals," the Survey states, "and the competition 
between books is a contributing cause to every major problem of the 
industry." Crowding too many titles into short periods of market time 
taxes the bookseller's capital and his display space. It also taxes the 
care and attention of everyone who touches the book publisher, job- 
ber, retailer, and book reviewer necessarily causing the neglect of 
some titles in favor of others. "Of the total number of trade books 
available," says Cheney, "not sixty per cent receive even fair merchan- 
dising attention." The ideal remedy would be to publish fewer titles. 
But who would know which titles to cut out? 

That is why the vision of the publishers sitting around a table and de- 
ciding to cut down their lists is almost equal, in stirring the hollow 
laugh, to a disarmament conference. But the picture is lugubrious 
compared with the one of each publisher, surrounded by his staff, doing 
the actual cutting down. 

Despite seventy years of marketing, book publishing is still an insu- 
lar industry. The argument of Gaynor's lecture is simple. The eco- 
nomic organization of the American book trade, still patterned on its 
nineteenth-century origins, includes hundreds of small business units, 
each handling many (and many of the same) products in limited quan- 
tities. The sale of these books (and often the books themselves) de- 
pends on the variable desires and predilections or the "taste" of 
thousands of groups of people; but none of these business entities has 
ever had the means to afford a thorough and systematic exploration of 
book markets. Anyway, nobody has ever analyzed public taste, except in 

Unable to find or to create any useful opinion-makers among the 
nation's book buyers, these individual enterprises have consequently 
looked to one another to define taste. By a process of trial and error, 
the publisher produces as many books as possible, testing the appeal 

The Clay Feet of Polk & Franklin 73 

of his products to other business entities within the trade, even as they 
pass these products along to the ultimate buyers. The rising costs of 
operating a business, and the high price of experimenting, have rein- 
forced this tendency to look within the trade for means of measuring 
the appeal of books. 

Gaynor's forthcoming lecture is all about this inversion of the book 
trade, and he illustrates his argument by describing three conditions : 
the plight of the retail bookseller, the ridiculousness of the publisher's 
advertising, and the problem of book-club distribution. Gaynor's argu- 
ment is so impressive and so discouraging that he himself sees more 
frightening implications than he expected. 

Retail bookselling, Gaynor's lecture points out, offers as timely a 
subject now as it did in 1872, when Frederick Leypoldt founded Pub- 
lishers' Weekly. In that year the United States, with a population of 
40,000 people, had about 3,000 book stores. Today, with a national 
population more than four times larger, there are fewer than 3,000 
retail book stores in the United States. 

Even this startling comparison falls short of the facts. Polk & Frank- 
lin's mailing list, for instance, includes only 2,100 stores whose gross 
receipts come primarily from books; and only twelve per cent of the 
names on P&F's list stores located in New York City, Philadelphia, 
New England, Chicago, St. Louis and scattered West Coast cities 
stock more than 2,000 current titles. (Gaynor does not express his pri- 
vate opinion that P&F cannot rely on any more than fifty of these stores 
for aggressive and intelligent marketing of current and backlist books.) 

Although the reprint publishers have discovered 100,000 new outlets, 
including cigar stores, bus terminals and supermarkets, most retail sell- 
ing of original hard-cover books is confined to a very few bookstores. 
The plight of the rest of the average booksellers has its effect on the 
entire trade. Gaynor emphasizes this with some of Jacques Barzun's 
observations in The New Leader (May 13, 1957), about the bookstore 
as a bottleneck "a plugged-up medicine dropper." 

The bookseller is caught between the rent of his poky shop, the high 
cost of shipping and accounting, and the vagueness of the public mind 
the mind which would forget the name Lucky Strike in a week if 

74 The Denatured Novel 

unprqmpted by ads. In this literal and figurative box, a bookseller 
would be a fool to stock anything but best-sellers and college diction- 
aires. Which means that the trade's effort at marketing concentrates 
on persuading him that some new work will be a best-seller, 

Surely, the bookseller offers no vital communication between the 
publisher and the public he seeks. Readers dependent on the retail 
trade often never see or hear of books they might otherwise buy, yet 
the myth of "the infallible advance sale" persists. His sales are negli- 
gible, yet his store has traditionally played a crucial part in the pub- 
lisher's planning. As an outlet for advance sales, it has long been a 
kind of laboratory for a book's dry-run. The transactions between pub- 
lishers' salesmen and the booksellers they visit have usually determined 
how and to what extent the publisher will advertise his book. To the 
extent that publishers have heeded the advance sale, they have acceded 
to the judgment of their salesmen and the booksellers. 

As further evidence of the trade's inversion Gaynor cites its methods 
of advertising, which he compares with other industries. In the national 
economy advertising has become the primary selling tool, the means 
of creating desires in order to sell products to satisfy them. The pur- 
pose of book advertising, accordingly, would seem to be increased sales, 
but publishers advertise only deviously toward this end. To be sure, 
every publisher goes through the motions of advertising books to con- 
sumers, but most of the media cannot reach his markets for prices the 
publisher can afford. 

Publishers buy no time on television or radio. They buy space in less 
than one per cent of the nation's 1,750 daily newspapers, in about five 
per cent of the nation's six hundred consumer magazines and in all 
trade and general book-review journals. But the rising cost of adver- 
tising space, beyond the increase in book prices, has reduced even this 
pitiful effort. (Gaynor remembers a memo he sent to Loesser about 
Red Beach Assault- "We can nicely afford a one-inch ad stating 'Com- 
pliments of a Friend/") And what the publisher cannot afford he 
tries to get for nothing through the efforts of his publicity department. 

Instead of telling more people about his books, the publisher in- 
vests in brand advertising of the most inverted sort. He has decided 

The Clay Feet of Polk & Franklin 75 

he can sell more books by advertising himself to the trade. He tries to 
create an image of his list to authors and their agents, to the editors 
of book reviews, to book clubs, to the story departments of motion pic- 
ture studios and to other publishers, in order to convince them of his 
strength and of the sales appeal of the books he sponsors. This, says 
Gaynor, is the tip-off. His customers are not readers. They are the 
authors of new manuscripts and the marketers who will sell the books 
he makes of them. In this sense his advertising moves in a closed sys- 
tem. It is meant first for the trade, which provides and buys titles, then 
for the consumers who buy books and other furniture, but only cas- 
ually for readers. 

There are some other consequences of this tight little book world. 
Advertising agencies assign and prepare most of this advertising. But 
of twenty-seven New York advertising agencies specializing in books, 
six or eight handle most of the publishers' accounts, a few of them rep- 
resenting at one time as many as twenty different publishers. In any 
other business this friendly little custom would constitute malpractice, 
particularly, as in the book trade, if all the agent's accounts wanted to 
create the same public image. 

Even when all the products are supposedly different one of the 
book trade's half-truths the agent must represent each book on its 
own terms, yet suggest its relevance to the common image of strength, 
integrity, discriminating taste and sales appeal which every sponsor 
happens to want for his own. It is rather like playing all the hands in 
a poker game and playing them all to win. Were advertising books 
merely a matter of buying space and writing copy, the agency would 
never earn its fees. But because books belong to publishers' lists, and 
because building the list and advertising it usually go together, the 
agency often consults on editorial matters. 

Such consultation occurs in those seasons of doubt prior to the pub- 
lisher's semi-annual sales meetings when, for any of a dozen vexing 
reasons, the list is not "complete." An author has not finished a manu- 
script, the possibility of libel has temporarily delayed a book, the dis- 
covery of new material necessitates the revision of another, or a novel 
in galley form may suddenly reveal everything wrong with its story. 
But the publisher must "fill" his list. "Filling" the list is poor publish- 

j6 The Denatured Novel 

ing. It ignores the quality of an individual book, but it constantly 
tempts every publisher. 

Remembering his fixed overhead and the uncertain market, he be- 
lieves he must produce even over-produce to compensate for the 
books now scratched from his list. He recalls rejected manuscripts from 
literary agents, detours a title or two from the mail-order department, 
and hurriedly hunts for a juvenile or even a technical book to "balance" 
the list. Without time to establish an advance sale he scarcely knows 
these new books, yet he must advertise them. The publisher obviously 
needs some coordination between list-planning and advertising, so he 
deputizes his advertising agent. 

Chances are, the agent can estimate more accurately than his client 
any new book's possibilities. Without the pressure of sales quotas he 
can judge a book more effectively on its own terms. In this seasonal 
breach of publishing routine the advertising agent becomes more than 
ever a taste-maker. Even more directly than the publisher's salesmen 
or the bookseller, he shares the editorial responsibility. 

Direct-by-mail advertising is a more reliable means of communicat- 
ing with the book buyer because it aims at sales rather than customers. 
Although the results are limited, they are immediate. The responses 
from a selected mailing list to the announcement of a prospective book 
are so reliable that the publisher can accurately forecast its sale. By 
split-run advertising mailing different announcements of the same 
book he can even learn how to increase sales by varying that book. 
(Gaynor might have added that Polk & Franklin often builds a book to 
order in this fashion, starting with nothing but a title, a description and 
a mailing list. But this is beside the point Gaynor means to make.) 

Selling books by mail, in the form of the book club, has revolution- 
ized trade-book marketing. But Gaynor calls this merely one more 
symptom of the trade's inability to communicate with readers. The 
book club sells a contract which obligates a purchaser to buy books in 
return for a bargain. The book which a club sells is usually a simul- 
taneous reprint of a trade edition. The large club rents plates from the 
publisher, manufactures its own books, and pays royalties jointly to 
author and publisher. Its prime costs are less than the publisher's. Plate 
rental eliminates most of the expense of composition, and the large dis- 

The Clay Feet of Polk & Franklin 77 

tribution a club can guarantee reduces printing and binding costs. 

The American archetype of this business, the Book-of-the-Month 
Club, has exploited these factors since its founding in 1926. During its 
first year, even without dividends and with readers paying postage, 
this club managed to distribute over 230,000 books to nearly 4,700 
buyers. The membership reached 110,000 by 1929 but shrank with the 
Depression, so the Club inaugurated the book-dividend system in 1931. 
In its twentieth anniversary year the Book-of-the-Month Club dis- 
tributed more than 11,000,000 books to about 900,000 members. 

On this pattern, many hundreds of book clubs have been founded 
with nothing but small capital (or credit) and a mailing list. Many 
have gone bankrupt or been absorbed by others, and new ones con- 
tinually appear. Among the giants in the field by 1950, the Book-of- 
the-Month Club and the Literary Guild, and a third, the Dollar Book 
Club, accounted for eighty per cent of all current book club sales. The 
book clubs are the publishers' best customers. They pay more than half 
of all trade publishers' subsidiary income. Their advertising increases 
the trade sales of books by as much as ten to fifty times. 

The effect of book clubs upon publishiruj has caused a lengthy and 
unresolved debate. By selling large quantitiefc x of relatively few current 
titles they have exploited best-sellerism; they have made a few authors 
rich, but they have jeopardized the retail booksellers in sum, trade 
publishing as now practiced could not exist without them. But one of 
their characteristics, says Gaynor, has been too long overlooked. Not- 
withstanding their enormous sales even because of them the book 
clubs have made even more distinct the difference between the book 
buyer and the reader. 

Gaynor refers to the transaction between the club and its members. 
The critical, unpublicized fact of all these clubs is the tremendous 
turnover in membership they have had to sustain. About forty per cent 
of the Literary Guild's membership accepts the Guild's monthly selec- 
tions. Offering a more varied selection, the Book-of-the-Month Club 
manages to sell books to a monthly average of only about twenty-five 
per cent of its members, and half of its membership changes every year. 
The significance of this high mortality rate, Gaynor believes, is re- 
vealed in book-club advertising. 

78 The Denatured Novel 

The clubs advertise a bargain: books at less than retail price plus 
free book-dividends. The Book-of-the-Month Club spends millions of 
dollars each year on membership campaigns merely to maintain its 
status quo. But neither the cancellations of membership nor the pur- 
chases of books in the first place have much to do with reading. 

People will accept almost anything free, even books, Some will con- 
tinue to buy book bargains until their shelves are full; others women 
mostly even intend to read them. But with astonishing regularity new 
books arrive, adding to the pile of homework. The self-improvement 
campaign which once looked so promising begins to pall; with guilt 
and annoyance the purchaser cancels her subscription. What happens, 
Gaynor explains, is that book clubs continually oversell their selec- 
tions, offering give-away books to people who would not normally read 
them or even buy them. In so doing the clubs contradict the first com- 
mandment of the trade: "Thou shalt not oversell." 

Changes in book-club memberships within the past ten years have 
borne this out. By 1950 the Book-of-the-Month Club had lost nearly 
half the number of subscribers it had four years earlier. Also in 1950 
the Reader's Digest Condensed Book Club began with 200,000 mem- 
bers; it now has nearly 2,000,000. Gaynor pauses over this comparison. 
The Reader's Digest Condensed Book Club, as its name implies, sells a 
different product. 

Ironically for the trade publisher the demand for condensation of 
original works appears to be based soundly on readers' habits and de- 
sires. This is far more secure than the appeal of the give-away clubs. Its 
success poses some sobering questions about reading in the United 
States, a subject which suggests exploration if book publishing is to 

The condensed book club offers a better bargain than any other in its 
promise of entertainment with less effort. It. sells books that have been 
pre-read, with the difficulties of reading thoughtfully removed a mas- 
ticated product that is the closest correlation yet discovered between 
book publishing and American readership. Perhaps the perfect books 
for this democratic literacy will turn out to be only Tables of Contents. 

Gaynor's lecture stops short of asserting the reason for this. By over- 
producing books which are subject to the discrimination of book mar- 

The Clay Feet of Polk 6 Franklin 79 

keters the trade publisher has lost control of the whole operation. He 
has lost control to his customers, and these customers are not readers 
or even book buyers. They are the book marketers. 


In the light of the trade's colossal disregard of the Cheney Survey, 
Gaynor can expect no more than a polite acknowledgment of his lec- 
ture. Publishers' Weekly will applaud him in an editorial statement. 
But even Polk & Franklin will continue its publication of trade titles 
for subsidiary markets, its investment in a larger textbook department, 
and its production of manuals of this-and-that for direct-mail customers 
(in spite of the fact that Gaynor would rather publish any competent 
novel than the "Song of Solomon" profusely illustrated). 

Basically a distributor of books, the marketer originally had no edi- 
torial function. But the markets for fiction have changed in the twen- 
tieth century and grown far beyond the bookstore clientele. Marketers 
have persuaded the trade publisher to change his wares. Gaynor's tact- 
ful omission about book clubs comes close to the point. There may in 
fact be no editorial collusion prior to a club's purchase of a book, but 
there need be none, so long as the publisher remains aware of the kind 
of purchases a club has previously made. "Completing" the list is often 
the trade's euphemism for loading it with candidates for subsidiary 

The so-called subsidiary markets now support the trade publisher. 
By a slip of the bookkeeper's pen "subsidiary" has become "subsidy." 
Ever since the iSgo's, trade publishing has been continuously involved 
with one or several of them; not groups of individual purchasers, but 
other media which use fiction. The earliest of these markets were the 
magazines which existed to carry advertising to national circulations. 
Then the movies effectively influenced the book trade after the U. S. 
Supreme Court decreed copyright protection to authors and publishers 
for filmed adaptations of their books. Most recently the inexpensive 
paperbacks have revolutionized the sale of fiction. 

80 The Denatured Novel 

Gaynor's lecture set out to illustrate the book trade's concentric in- 
fluence, but it concludes in dismay over the success of appealing to 
readers who don't want to bother with the effort of reading. Having 
worked it through, Gaynor now reads his own message with awe. Surely 
the problem has outgrown Polk & Franklin. Some supra-agency must 
take over. Gaynor has just decided to do what you might have expected. 
He has enclosed a pre-print of his lecture in a long letter addressed to a 
large philanthropic foundation (the Maxwell-Essex Fund) requesting 
several million dollars for the book trade to study the acute problems of 
mass literacy. 

Chapter IV 

Slick Nirvana 

POPULAR fiction has always assured the reader about himself and his 
possibilities that humanity is naturally good, that the individual 
counts for something, that he can improve and that material success 
has Divine benediction. When the novel is on the right track with 
these ideas it can show all the world's wickedness with impunity. In- 
deed, the assurance of better things to come does not have to be in- 
sistent. So long as it is there at all, the reader will magnify it as needed, 
while enjoying the trials and crises of righteousness. 

Fiction that makes a policy of this particular life-assurance will al- 
ways find readers, so long as it can be distributed to them, which is why 
the commercial magazines have had so much to do with the develop- 
ment of the American novel. By a plurality of hundreds of millions of 
readers, Americans have read more fiction in magazines than in books. 

Particularly since the iSgo's, magazines have financially supported 
most American writers. That eminent magazinist, William Dean 
Howells, spoke confidently of this fact in October 1893, in his epochal 
essay 'The Man of Letters as a Man of Business." As he told the readers 
of Scribner's Magazine, "story-telling is now a fairly recognized trade, 
and the story-teller has a money-standing in the economic world." 
Authors could now live, he said, "and live prettily enough, by the sale of 
the serial publication of their writings to the magazines." 

Howells included more than just the few or famous writers: "a much 
larger number of clever people who are as yet known chiefly to the edi- 
tors, and who may never make themselves public, but who do well at a 


8z The Denatured Novel 

kind of acceptable work." In fact, Howells announced, "the prosperity 
of the magazines has given a whole class existence which, as a class, 
was wholly unknown among us before the Civil War." 

Howells was talking about the cumulative effect of many changes. 
Although installment publishing had occasionally occurred since the 
eighteenth century, it became popular when Harper's New Monthly 
Magazine made it a policy after 1850. The editor of this same maga- 
zine, and a novelist himself, Howells could talk with authority about 
installment fiction. What made his statement newsworthy was the 
tremendous growth of the popular magazines after the i88o's, both in 
numbers and in circulation. 

The growth of the magazines coincided with the book trade's organ- 
ized promotion of literature which followed the new Copyright Act of 
1891. In fact, the Federal Government had already (incidentally but 
immeasurably) assisted popular literature by subsidizing the maga- 
zines. An Act of Congress in 1879 which reclassified the mails desig- 
nated magazines as second-class matter, subject to the privilege of a 
new and lower mailing rate. When the native market for fiction began 
to develop, therefore, the magazines were the chief means of reaching 

In A History of American Magazines: 1885-1905 (1957) Frank 
Luther Mott estimated that the number of magazines in the United 
States increased from 3,500 to 6,000 during the twenty-year period of 
his study, and that some 7,500 new magazines had been started. There 
was more to what he called the "Magazine Revolution" than simply 
growth in numbers. The invention of the halftone process, of adapting 
the photograph to an engraved printing plate, made lavish illustration 
commercially possible. This changed the appearance and the entire 
nature of magazines. 

The new ten-cent magazines in the iSgo's were handsomely illus- 
trated and printed on slick (coated) paper devised for photo-engrav- 
ing. Their prosperous appearance amplified their contents. They were 
timely, favoring contemporary subjects, in fiction and in factual ar- 
ticles, for the entire family. They offered entertainment, education, 
romance and vicarious excitement to as large a public as possible. They 
were popular in every sense of the word. 

Slick Nirvana 83 

Subsidized by low mailing rates, the publisher could afford to build 
the circulation of his magazine. If he could guarantee a large subscrip- 
tion list, he had something far more valuable than the income from his 
subscribers. He had a possible market for anyone with a product to 
sell. So he became a broker in names, a dealer in consumers, to whom 
manufacturers for a price could advertise their products. 

Finley Peter Dunne's spokesman, Mr. Dooley, once commented on 
the magazines 7 economic incentive. In American Magazine (October 
1909) he found fault. There were not enough ads to suit him: "What 
I object to is whin I pay ten or fifteen cents f r a magazine expectin' 
to spind me avenin' improvin' me mind -with th' latest thoughts in 
advertisin' to find more thin a quarther of th' whole book devoted to 
lithrachoor." Some publisher, he hoped, would be bold enough to 
devote his magazine entirely to advertising. Since Dunne was both a 
magazine writer and an editor, his Mr. Dooley carried some weight 
with his opinion that "Lithrachoor an' business are hooked up to- 

Advertising rates depend on the magazine's circulation, so the pub- 
lisher wants to build and maintain as large a readership as possible. His 
surest way is to capitalize on established interests among his readers. 
The editorial content of his magazine is a means to this end utterly 

There are many publics, which is to say many large groups of con- 
sumers sharing common interests. Every magazine cultivates some such 
group; ideally, it is a unique editorial package which consolidates and 
encourages the habits of its readers. The literature which serves it must 
remain dependably the same. 

To insure his own stability the magazine publisher seeks a variety 
of publics at the same time, but not for the same magazine. He pub- 
lishes several at once. For many years the Curtis Publishing Company 
issued three very different slicks: Ladies Home Journal (1883-) a 
women's service magazine; Saturday Evening Post (1821-), a weekly 
for the men of the family; and a farm magazine, Country Gentlemen 
(191 1-1955). The Crowell Corporation (later Crowell-Collier) bought 
and issued Womans Home Companion (1897-1957); Farm and Fire- 
side (1906-1929), which became Country Home (1929-1939); and 

84 The Denatured Novel 

Collier's (1911-1957), a weekly family magazine to compete with 
Curtis' Saturday Evening Post. The Hearst publishing enterprise 
moved into the magazine field with products similarly compart- 
mented: two women's service magazines, Good Housekeeping (1885-) 
and Pictorial Review (1899-1939), and two of more general interest: 
Cosmopolitan (1886-) and Hearst's International (1901-1925), which 
Cosmopolitan later absorbed. 

In his attack on the vested interests in literature, Money Writes! 
(1927), Upton Sinclair decided in a chapter called "Artificial Selec- 
tion" that these chains were all "run exactly like the department 
stores and shoe-factory chains, upon the same principles of standardi- 
zation and mass production." But the size of a publisher's organization 
is incidental to the compartmentalization of his products. He is a 
specialist in audiences, because the advertisers must have an accurate 
gauge of their markets. For instance, McCall's (1897-) and Redbook 
(1903-) both published by McCall, go to two different groups of 
women, differentiated by the amount of the husband's income. The 
advertisements in McCall's, like those in Ladies Home Journal, appeal 
to a higher standard of living and reading. Redbook 9 s competitor used 
to be Crowell-Collier's American Magazine (1888-1956), but is now 
Hearst's Cosmopolitan. 

Nothing is left to chance. Conde Nast's House and Garden (1901-) 
and Hearst's House Beautiful (1896-) competitively seek an audience 
of similar interests but higher income than Nast's Better Homes 
and Gardens (1924-). Similarly, Nast's Vogue (1892-) represents fash- 
ions in women's wear differently than Nast's Glamour (1939-) or 
Street and Smith's Mademoiselle (1935-). And Hearst's Harper's 
Bazaar (1867-) goes to readers older and richer than readers of Seven- 
teen ( 1944-) , published by this same house. 

Aimed at a precise audience, the magazine must continue to gratify 
it with the same fare. In token of the publishers' strenuous attempts 
to stay'with the readers, most of the slicks have changed hands or been 
absorbed by others. But it is striking that despite the high mortality of 
the magazines themselves the literary product has remained constant. 

The editorial content of some of the failures, for instance, is repre- 
sentative of the entire slick field over the years, both in quality and in 

Slick Nirvana 85 

quantity. Woman's Home Companion (1873-1957), for instance, al- 
ways offered a rich fare in fiction, and counted among its contributors 
Mary E. Wilkins, Margaret Deland, Sherwood Anderson, Willa 
Gather, Booth Tarkington, Pearl Buck, Ellen Glasgow, Sinclair Lewis, 
John Galsworthy and Arnold Bennett. But the staple fare in each issue, 
beginning in the 1920*5, was the serialized novel (usually two) by 
women who could repeat their performances: Kathleen Norris, Edna 
Ferber, Dorothy Canfield Fisher, Mary Roberts Rinehart, Faith Bald- 
win and Taylor Caldwell. 

Another magazine which failed, Pictorial Review (1899-1939), 
offered comparable fiction by well-established authors: Edith Whar- 
ton, Carl Sandburg, Joseph Conrad, Emil Ludwig, Booth Tarkington, 
Gertrude Atherton; and characteristically the best of their writing. The 
pages of Collier's (1911-1957) tell a similar story. This magazine 
sought both British and American novelists, and built its circulation 
on big names. The roll call of Collier's contributors from 1900 through 
the 1930*5 correlates remarkably with the authors of best-selling novels. 

The life of a slick magazine depends on its rapport with its readers, 
which doubtless explains why these three magazines lasted as long as 
they did. And maintaining rapport is a delicate business. Bernard De- 
Voto, another eminent magazinist, described (in Saturday Review, 
October 9, 1937) some of the demands of this readership. "People do 
not read the slicks/' he said, "to encounter the brutalities, the pro- 
fundities or the complexities of experience. They read to have their 
ideas confirmed and their emotions ratified, to have their phantasy 
life stimulated, and to increase their knowledge of the minor sanctions 
and rituals of society but first of all they want to be amused." 

No writer can escape the reader's influence even the established 
and calculable writer. Temple Bailey's long and profitable career of 
writing magazine serials ought to make her an expert on what she calls 
"the idealistic presentation of young love"; yet she long ago developed 
the habit of writing a serial concurrently with its publication so that 
the readers' mail could help her determine the plot. For this agility, 
even in the Depression years, McCalfs paid her $60,000 a serial. 

The syndrome works both ways. The magazine also informs the 
readers of their own habits and appetites. The editor is a taste-maker, 

86 The Denatured Novel 

defining and stimulating the habits, desires and impulses of his readers 
simply by choosing and shaping what they read, 

So the editor uses his prerogatives in telling authors what he wants. 
Even Faith Baldwin insists on this. Contributing to a symposium 
called The Writer s Book (1950), she says, "The writer has something 
to sell; the editor something to buy. I doubt that it is ever a seller's 
market in this field. The buyer is in the better position; and writers are 
a dime a dozen/' Consequently, "An editor is prone even obliged 
to tell writers what he believes they should write/' 

Upton Sinclair put it differently in Money Writes! indicating 
the magazines' "Artificial Selection" of their material. "They know 
what they are going to want a year from now, and they order their 
stories as they order their trainloads of paper from the mills; they even 
order their writers; they will take a young genius and 'make' him, ex- 
actly as Lasky or Paramount will turn a manicure girl with pretty pout- 
ing lips into a world-famous 'star'." 

A more patient explanation of the editor's collaboration with his 
author is Elmer Davis' Bowker Lecture, "The Economics of Author- 
ship" (1940). Davis wrote: "When a magazine editor has located 
somebody who has that aptitude, and who can write well enough so 
that his name has, or can acquire, a circulation value, he has a pearl of 
great price; and he is going to feed that man ideas, when he no longer 
has ideas of his own." This goes beyond merely priming the pump. 
Margaret Culkin Banning, who wrote a serial a year for thirty-five 
years, makes this clear. In the symposium called The Writer's Book 
she says: "The story which the novelist intends primarily for serial sale 
and its rewards is too often the child of the editor. The novelist only 
has it committed to his care to bring up and develop." 

The editor's aim and the reader's, although compatible, are not the 
same. The fiction the editor prints serves the need of his sponsors, the 
advertisers, to the extent that it must capture and hold as many readers 
as possible. This is a restraining influence. The fiction may not offend 
business in general or in particular, or American government of any 
variety, or any religious, civic or military organization. The fiction may 
not contradict the mores of the average consumer-reader without pun- 
ishing the offender. These well-known prohibitions are similar to the 

Slick Nirvana 87 

ones levied on motion pictures; most of them also apply to the inex- 
pensive paperback books. The prohibitions and the devious ways 
around them belong not to art but to manufacture. 

Slick magazine fiction must entertain several million readers with- 
out offending them or straying beyond their interests. It must offer 
the same fare over and over again without seeming to. These require- 
ments prescribe the editorial content of every mass-circulation maga- 
zine. The most familiar evidence of this requisite product and of its 
effects on the economics of writing is undoubtedly the Saturday Even- 
ing Post. 

The Post contains more information about the national personality 
than any other publication in the world. It is the slick by which all 
others are measured. The slurs about its middle-brow superficiality, and 
the response that most of the disparaging writers could not meet the 
Post's standards of publication, are quite familiar. Charge and counter- 
charge are grounded to the man who guided this magazine to its emi- 
nence George Horace Lorimer, the Post's boss from 1899-1936. 

Lorimer's first editorial page (September 30, 1899) declared the 
Post's homage "to the great mass of intelligent people who make 
homes and love them, who chose good lives and live them, who seek 
friends and cherish them, who select the best recreations and enjoy 
them." The incense is unmistakable. The sweet dream in which "the 
great mass of intelligent people' 7 would share was already being fab- 

For nearly forty years the Post demonstrated what this patriarch 
meant. He was a conservative and a nationalist. Like his readers he 
believed in God, in love and marriage, in college educations, in success, 
in the conservation of wild life and in the policy of American isolation. 
He opposed the League of Nations and the New Deal; he fought for 
restrictive immigration, for free enteqDrise in business, and for the 
sanctity of the family. The Post amplified all of these convictions. The 
magazine was an extension of the man. 

Lorimer selected all of the material, wrote the editorials, selected 
the cover and the story illustrations, even censored the ads. He kept 
his fingers on his writers and on their pens. He frequently corre- 
sponded with them and entertained them in order to check on the 

88 The Denatured Novel 

progress of their work. Emerson Hough referred to his novel, The 
Covered Wagon (1922), as their "partnership book," in token of Lor- 
imer's proposing it in the first place, helping him to plan it, and con- 
sulting in the revision. Occasionally Lorimer even ghosted. When 
David Graham Phillips was murdered, his unfinished serial, "Grains 
of Sand" ( 191 1 ), was running in the Post; Lorimer finished the manu- 

Lorimer paid his authors handsomely, and set the rates for all other 
slicks to beat; but paternalism went with the pay. He would not bar- 
gain over manuscripts or offer contracts, but after an author had pub- 
lished in the Post, Lorimer took his loyalty for granted and assumed 
that the Post would have first refusal on the next story/Leaving the 
Post for another magazine as Irvin S. Cobb did, and Jack London, 
Ring Lardner and Peter B. Kyne Lorimer considered a breach of 
friendship. Such paternalism, along with Lorimer's shrewd sense of 
fiction and vigorous scouting, secured an imposing list of contributors. 

Some of these might surprise you: Stephen Crane, Frank Norris, 
Theodore Dreiser, James Branch Cabell, Edith Wharton, Willa 
Gather, Ellen Glasgow, Zona Gale, William Faulkner and Rebecca 
West. Others you might expect: Margaret Culkin Banning, Katherine 
Brush, Temple Bailey, Edna Ferber, Fannie Hurst, and Fanny Heaslip 
Lea and Mary Roberts Rinehart. Yet they all published in the Post, 
and Lorimer made no distinctions. 

He had, in fact, a narrow view of the word "literature," saying that 
only time could decide literary status, and that he did not buy manu- 
scripts for their promise of future distinction. An issue containing a 
Dreiser story and a serial by Temple Bailey offered no problem at all, 
as long as each of them was good enough to be published. 

The variety, however, was more apparent than real, as indicated by 
the frequency with which these writers published in the Post. A few 
of them published in it all of the time. This observation bears on the 
economics of magazine authorship. The more often a writer's work ap- 
pears in the pages of a magazine, the more it is worth in building that 
magazine's circulation, for the reader gets in the habit of looking for 
it. Lorimer paid everyone well, but he paid most to the competent 
writer who could repeat himself. 

Slick Nirvana 89 

The pages of the Post during 1934, tne ' ast Y 631 * Lorimer's presi- 
dency (before he became board chairman), show how well he had ex- 
ploited the value of repetition. During that year each issue, of about a 
hundred pages, carried two serials, four or five short stories, and four 
or five articles. Among the contributors Sophie Kerr, Mary Roberts 
Rinehart, E. Phillips Oppenheimer, Oliver La Farge, I. A. R. Wylie 
and Kenneth Roberts each published a serial; Margaret Culkin Ban- 
ning, a serial and two articles. John Marquand published a five-part 
serial, a six-part serial, and five short stories; Walter D. Edmonds con- 
tributed a serial and four short stories. There were also nine stories by 
Everett Rhodes Castle, six by Booth Tarkington, five by Richard 
Macaulay, four each by Thomas Beer, Ben Ames Williams, Joseph 
Hergesheimer, Lucian Gary and Corey Ford, and three by Octavus 
Roy Cohen, William Faulkner, and Bernard DeVoto. 

Magazine publishing thrives on over-production; it contradicts the 
familiar law of supply and demand. The more a given editorial product 
appears in the magazine, the more valuable it is. The familiarity of the 
author's name helps to maintain circulation. In his Bowker Lecture 
Elmer Davis explained this fact. "If you sell any given magazine ten 
stories in a year," he said, "you will get more for each story than if you 
sold them only three stories a year. It may well be that three stories, 
three good stories, are all you have in you that year; but if your produc- 
tion falls off your price is likely, sooner or later, to fall off too." 

The writer's alternative is to force his production beyond what he 
has to say. "So you write your three good stories; and then you use the 
tricks of the trade that you have learned and your knowledge of the 
tastes of the editors of that particular magazine, to knock off seven 
more stories that are good enough good enough to sell with the ad- 
vantage of a name well known to the magazine's readers, even though 
they might be sent back if they came up from Joe Blotz of Podunk 

The conditions of the industry which encourage this enormous over- 
production penalize all kinds of writers. "But the most regrettable 
case of all, which unfortunately is far too common in this country, is 
that of the man who has something to say, but not much." The mag- 
azines' demands for more of the same "make it practically impossible 

90 The Denatured Novel 

for a writer to retire into dignified silence when he has said all he has 
to say, or to lie fallow for a while if he is going through a period of 

The predicament of success is a peculiar condition of American au- 
thors. For most of them this predicament consists of publishing in the 
magazines to pay their bills while trying to write, as the saying goes, 
the Great American Novel. This has urged many commercially suc- 
cessful writers to a pathological hatred of the magazines. 

Henry James's tales of writers who struggle between conscience and 
commerce dramatize a universal condition of the artist. But one way 
or the other his writers always choose between the alternatives or fail 
in attempting both. James's stories are not parochial enough to fit 
the American scene; the really genuine American fable is Jack London's 
novel, Martin Eden (1909), about a writer who ironically succeeds in 
serving both masters. 

Martin Eden wants to write a masterpiece, to correlate all knowledge 
into the theory of one book; but to keep from starving meanwhile, he 
grinds out stories for the magazines and newspaper syndicates. So he 
works out the perfect formula. He finds "that the newspaper storiette 
should never be tragic, should never end unhappily, and should never 
contain beauty of language, subtlety of thought, nor real delicacy of 
sentiment." But it must contain sentiment, "plenty of it, pure and 
noble . . . Tor-God-my-country-and-the-Czar' and 'I-may-be-poor-but- 
I-am-honest' brand of sentiment." 

This is what will gratify the reader. The mechanical contrivance is 
simple, 'The formula consists of three parts: (i) a pair of lovers are 
jarred apart; ( 2 ) by some deed or event they are united; ( 3 ) marriage 
bells. The third part is an unvarying quantity, but the first and second 
parts can be varied an infinite number of times." 

Eden's problems really begin after he has succeeded in both his en- 
deavors. His masterpiece, a philosophical essay as unpopular as you 
could imagine, becomes a great commercial success for all the wrong 
and whimsical reasons. It brings him immense notoriety (there is a 
fine deadpan satire of his publisher's antics); and the magazines fall 
all over themselves begging Eden for the reams of hackwork they have 
already rejected. 

Slick Nirvana 91 

This whole experience, however, has broken him. He has rendered 
his life into his masterpiece and has nothing of his own left to say, and 
he has outgrown the need to create the dreams of his formula-days. 
His situation is hopeless; and since he has nothing left to be, Jack 
London puts him to death. The suicide's failure to resolve anything is 
only one of the difficulties of this novel. Nevertheless, the condition 
of Martin Eden-the-writer, which Jack London abstracted from his 
own life, is a fiction which sounds like a good many biographies. 

The writing career of Stephen Vincent Benet was just such a battle 
with success. For nearly twenty-five years Benet had to depend on the 
slicks for most of his income. While his agent, Carl Brandt, master- 
minded a career for him with his Cinderella stories, Benet was always 
trying to get one story ahead so that he could explore the American 
past and write poetry about it. He desperately needed the magazines in 
order to live, and he hated them for it. 

Benet's diary and his correspondence record his continuous battle 
with the magazine editors "ravens/' he called them over their re- 
quirements for the "candy-laxative" product: their demands for a 
happy ending or a less controversial story. Brandt's rationalization, 
"You can change it in the book, 7 ' only maddened him more. The irony 
of Benet's first popular success came close to Martin Eden's. He did 
manage to write what he wanted, and John Browns Body (1928) be- 
came a convincing success. Its large trade sale and Book-of-the-Month 
Club distribution brought Benet money and fame. But the truth of 
Benet's life was more impossible than London's fiction. 

The stock market crash of 1929 sent Benet back to the magazine 
chain, and only after he published "The Devil and Daniel Webster" 
(in the Post in 1936) did he find some small liberation. It was just a 
parole, however. His prices were higher, and he could sell a little more 
of what he wanted to write. His disdain of the magazines lasted until 
he wrote himself to death. 

F. Scott Fitzgerald's disparagement of the slicks is well known. His 
letters in The Crack-Up (1945) show how critical he was of his own 
work which the slicks published. It was typical of his feelings that he 
referred Edmund Wilson to the book version of Tender Is the Night 
(1934) rather than the magazine version which "in spots was hastily 

92 The Denatured Novel 

put together." He had complained to John Peale Bishop that although 
he was getting $2,000 a story, they grew worse and worse, and his am- 
bition was to be able to ignore the magazines and write novels. Once, 
when he could afford to, he turned down an offer of $15,000 for the 
serial rights of The Great Gatsby ( 1925 ) . 

Even Booth Tarkington, who sold all the fiction he ever published 
to the magazines, took the same dim view of them. When his young 
friend and neighbor, Kenneth Roberts, learned that Lorimer had in- 
structed him to make drastic cuts in his manuscript of The Lively Lady 
(1931) he asked Tarkington's advice and then wrote in his diary: 
"Consulted Booth; he said at once that I should agree: that magazine 
publication was a purely temporary affair, forgotten in a matter of 
weeks. Only the book lives and is remembered." 

Roberts included this in his personal record, I Wanted to Write 
(1949), and among these odds and ends of a long career were some of 
Tarkington's letters which related to this magazine-and-book dilemma. 
Tarkington was impatient with the novel reader who "seeks to escape 
from life itself by the reading of romances a very numerous type." 
Though she does ask that the characters be plausible enough for her 
to accept them while she reads, "she doesn't want an interpretation of 

... it may happen that a serious writer writes a book that just by 
chance satisfies and pleases this enormous class of readers; but it prob- 
ably won't happen often, since, being serious ... he can't please them 
any more than their own lives do, since what he's doing is trying to 
paint life as truthfully as he can, and not to tell children bedtime 

The double standard of magazine fiction and book fiction was so well 
established in the trade that when John Marquand was ready to pub- 
lish The Late George Apley ( 1937) his publisher advised him to adopt 
a pseudonym so as not to injure his profitable reputation in the slicks. 
This condescending to serious fiction so annoyed Marquand that he 
determined to set the record straight. "I have never been able to be 

Slick Nirvana 93 

patient with this sort of literary calcification/' he explained. "I have 
never understood why a sinner is not allowed at least to attempt re- 
formation in the American world of letters." 

He wrote this in one of the prefaces in North of Grand Central 
(1956), a trilogy of his first New England novels. After twelve years 
of writing for the slicks Marquand became tired of their inhibitions, 
and since he could afford to he began to write to suit himself. With 
The Late George Apley he made a startling discovery: 

I found that the task of writing it, though occasionally difficult and dis- 
couraging, was far easier than the effort I had previously expended on 
obviously mediocre serial stories. Yet the result was apparently better 
than anything I had achieved previously. 

The reason for this was so obvious that I should have recognized it 
years before. For almost the first time in my life I had written about 
something that I thoroughly understood. 

He had not been able to write from his own experience for the maga- 
zines; as a professional he had trained himself to write as "an outsider/' 
to offer that deft touch of "reality in the details" to satisfy the casual 
reader. But his "contrived entertainment" had nothing to do with his 
own experience, without which an author has nothing worth saying. 
Without the original experience, "no matter how great may be his 
skill and brilliance, he is a huckster who is trafficking in the inflated 
currency of artificiality." 

Marquand's discovery about himself casts light on the awful shadow 
of success in which so many American writers have found themselves. 
A man must write what he knows and feels, or he cannot believe in 
what he writes. If the author cannot believe in it, who will? Like any 
other fiction, slick fiction involves this matter of belief. The ideal slick 
story is far more than the arrangement of a conflict which resolves 

For enough money almost any professional writer can learn how to 
use a few conventionally off-beat details to suggest a character, how 
to make a scene look authenic, or how to hide the evidence of a story's 

94 The Denatured Novel 

conclusion in the conflict itself. These are the techniques by which an 
author tells a slick story and gets away with it. But there is more to 
slick fiction than the correspondence courses can teach. It must con- 
vince the reader that what he wants to believe is so. 

The slick world is a dream-world with an ideology as old as original 
sin. In that world the reader is assured of happiness ever after. Most 
readers of the slicks are women, although women have no monopoly on 
wishful thinking. Everyone desperately wants to rationalize the miser- 
ies of the world, and to believe, even temporarily, in some plausible 
way of making it all better again. This is the easiest way of vindicating 
oneself. Any story which makes this heart-warming face-saving gospel 
seem believable is bound to be popular. 

This economic fact is of great importance to magazine publishers. 
The slick magazine is built on mass circulation, no matter what the 
social stratum of its audience. To keep it so, the editor will always try 
to gratify this great common yearning for human nature to be good 
and for goodness to win. 

The trick is to find the writer who believes in it enough to make it 
convincing. Frank Munsey, the originator of the ten-cent magazine, 
once exclaimed that good writing is as common as clam shells, but 
good stories are as rare as statesmanship. If an editor finds a writer 
who can write this rare story and write it often, money is no object; 
$60,000 a serial is well worth it. 

There are very few competent writers who believe that courage and 
loyalty will triumph over ignorance and evil, and that virtue will 
finally get the better of temptation. The triumph of goodness did not 
happen to be the truth of Scott Fitzgerald's or Stephen Benet's or 
even Martin Eden's experience. So the magazines must find writers 
who can counterfeit the dream-world convincingly and often. 

All the rule books say this cannot be done by patronizing the reader, 
and they are almost always right. An author cannot borrow a philoso- 
phy and make it convincing except occasionally, with great skill and 
good luck and a desperate need of money. The bare possibility that a 
writer may earn what he needs without contaminating himself in the 
process has made the dream a nightmare to all the Martin Edens. 

Slick Nirvana 95 

Sinclair Lewis has spoken for them all. In 1930 he became the 
first American to win the Nobel Prize for Literature, and no sooner 
had he won it than he carried to the world his quarrel with the slicks. 
Of all American writers, he mourned, "We still most revere the writers 
for the popular magazines who in a hearty and edifying chorus chant 
... the bucolic and Puritanic simplicity of Uncle Sam." This was 
partly a confessional, for Lewis himself was a thoroughly discontented 

Even before Main Street was published Lewis had worked his way 
up to $1,000 a story in the Post, yet he spoke, with a sense of compro- 
mise, of "earning a living by nimble dives into Saturday Evening Post." 
These nimble dives required all his dexterity; as he explained, "I have 
steadily sought to work out a means of doing as honest work as the 
powerful negations of the magazine editors would permit." He wrote 
this to Carl Van Doren in a letter dated November, 1920 (later pub- 
lished in his selected writings, The Man from Main Street, 1953). 

The pain of Lewis 7 compromise was acute during the two years in 
which he finished and published Main Street. His letters to his pub- 
lisher, Alfred Harcourt (in Harrison Smith's From Main Street to 
Stockholm, 1952) record the author's struggle with himself. After 
Harcourt, Brace announced the novel to the trade in May 1920, ad- 
vance orders promised an unusual sale; and the excitement of reviewers 
promised even more in the way of prestige for its author. But Lewis 
was out of money. While finishing the novel, he had managed to pay 
his bills by selling stories to the Post and to Harpers, but as yet without 
any book royalties he was worse than bankrupt. His future was compli- 
cated by the prospect of a succes d'estime whose sales, he felt, his mag- 
azine writing would only injure. He described his dilemma in a letter 
to Harcourt (November 30, 1920). 

I am, frankly, having a hell of a time in trying at once to turn myself 
back into the successful S. E. P. writer I was a year ago and yet do 
for them nothing but stories so honest that they will in no way get me 
back into magazine trickiness nor injure the M St. furore. 

96 The Denatured Novel 

He explained that he had just destroyed a sixty-thousand-word manu- 
script which he could have sold handsomely to the Post, "but which 
was so shallow, so unreal, so sentimental that (featured as they do 
feature a serial, even a short one) it would have been very bad for 
Main Street." 

God knows I don't expect you to bear the responsibility for this, which 
may have been foolhardy. I relate it only to prove how vigorously I 
have been attacking this problem. . . . I'm going, of course, to go on 
plugging at the Post, but I don't believe I shall ever again be the facile 
Post trickster I by God was for which, doubtless, we shall in the long 
run be glad. 

Lewis' experience with serializing his novels aggravated his either/or 
attitude about fiction. Free Air (1919) had only a modest sale, which 
he blamed on its manner of publication. The Post had run the orig- 
inal serial which Lewis then revised, nearly doubling its length for a 
book. Lorimer persuaded Lewis to run the new material as a separate 
serial in the Post. When the book failed Lewis decided its prospective 
buyers had already read it. 

He and Alfred Harcourt agreed thereafter to withhold the serial 
rights to his novels, but neither one lived up to this agreement. Har- 
court later sold the magazine rights to Arrowsmith ( 1925) for $50,000; 
and Lewis wrote a serial for Colliers which he persuaded Harcourt to 
bring out as a book. This was called Mantrap (1926). Neither author 
nor publisher regarded it highly, although Lewis felt called upon to 
defend it. He wrote to Harcourt (November 10, 1925), "I recall noth- 
ing shoddy in it, and as for the critics who insist that I have no right to 
do anything but social documents, they may all go to hell." 

Lewis' lame excuse emphasized Mantrap's curious relation to his 
other novels of the period. It followed Main Street (1920) and 
Babbitt (1922) and preceded Dodsworth (1929), all of which it ap- 
proximates and all of which exceed it. That it was not a "social docu- 
ment," as he put it, is one way of saying that it does not represent 
Lewis' view of experience. 

Slick Nirvana 97 

In fact, Mantrap offers no view of experience whatsoever; no presid- 
ing idea grows out of its conflict. It is a listless failure. Yet it is mechan- 
ically like the three contemporary novels which Lewis never intended 
to be serialized. A reading of the serial and novels reveals how Lewis 
tried to accommodate the magazine values in the serial. 

Mantrap tells of a mis-marriage between a fur trader at Hudson Bay 
and a manicurist he has met in Minneapolis. Mantrap Landing, the 
desolate scene of most of the novel, is inhabited by unfriendly Indians 
and friendly trappers, both of which jeopardize the virtue of this lonely 
bride. Their presence threatens her with a fate alleged to be worse than 
death. She loves her husband, but not enough to live with him faith- 
fully at Mantrap; and he loves her, but not enough to move away from 
the place. 

The marriage falls apart when a bachelor from New York City, who 
has arrived on a canoe trip, discovers his love for the woman and pru- 
dently decides to leave. But she escapes with him; and after an arduous 
chase across the wilderness the husband catches up with the pair, insist- 
ing that he must save his friend from his wife. They sit down and talk 
it over, but after exploring all possibilities they separate, each return- 
ing to the place he started from. Here the serial stops. No one has 
changed, and the story proves only the dissolution of a marriage which 
gave no promise in the first place. Nothing is really at stake because 
there is nothing by which to judge this marriage; the story contains no 
analogies. All its vigorous talk of a conflict between wilderness and 
civilization cannot fill the vacancy. 

The plot of this serial about the estrangement of a husband and wife 
is exactly like the plot of Main Street, of Babbitt, and of Dodsworth. 
The serial and the novels all depend on a scheme of trap and escape. 
In each, one spouse is trapped by the marriage, and the course of this 
marriage carries the meaning of the book. Babbitt's estrangement and 
reconciliation make him finally realize his own littleness in conforming 
to the mores which contradict what he really wants, Carol Kennicott's 
marriage makes her realize how in need of reform she herself is. Dods- 
worth's separation from his wife forces him to reappraise all the mate- 
rial values he has taken for granted. 

Mantrap has all of the machinery but none of the significance of 

98 The Denatured Novel 

these others. Lewis' belittling view of humanity, after all, was not con- 
genial to the dream world of the slicks. But when he tried to write a 
story without that belittling view he had nothing very convincing to 
put in its place. 

The scheme of trap and escape, so compatible with magazine fiction, 
continually tempted Lewis. Martin Arrowsmith's first fiancee and his 
second wife threatened to confine and distort his intellectual ideals. 
Lewis planned still another novel about the situation of a marriage, 
whose projection even though he never wrote the book seems per- 
fectly clear: "the story of a young couple bucking society in a city 
about Minneapolis; a story of that never yet adequately described but 
extremely important phase of American life middle-class existence in 
an American cross between town and city. . . ." 

He wrote this to Harcourt (October 22, 1919), continuing, "I am 
planning such a story, with a lot of drama and unexpectedness but also 
complete reality, as a serial for Sat Even Post, and I may do it before 
I go on with Main Street which will almost certainly NOT go as a 
serial." Here was the young writer counting two novels before they 
were hatched, calculating on the same plot, the same substance, but 
different values, and doling them out: one for George Horace Lorimer 
and one for Alfred Harcourt. 

Without that genuine belief in the dream world which distinguishes 
the real magazine serialist, each slick writer must find some way of his 
own to make the conflict turn out believably but also without offend- 
ing the readers' platitudes. This was where Lewis had failed. His maga- 
zine serial simply offered the logical consequence of what the conflict 
had promised at the beginning. Lewis had suppressed any adventitious 
change which might have made things turn out better. 

This was his concession to art avoiding the "magazine tricki- 
ness," as he put it. But in serving the magazine he also suppressed any 
idea the plot might have suggested as the plot had suggested ideas 
in his novels. Since Lewis never could write a good story, this compro- 
mise was not very convincing. 

The story sense, which Lewis lacked, was precisely what enabled 
Booth Tarkington to succeed in counterfeiting the values of the slick 
world. Tarkington had strong ideas, as Lewis did, about human petti- 

Slick Nirvana 99 

ness and irresponsibility and the degrading materialism America had 
come to worship. His convictions did not agree with the platitudes of 
the dream world, but he managed to make his stories assimilate his 
attitudes and still appear to turn out all right. 

Tarkington made the perfect compromise over and over again, the 
compromise that most novelists have had to face. He said exactly what 
he wanted to say and made everybody love it. For nearly fifty years his 
fiction about American families at home and abroad so epitomized the 
national culture that he became a celebrity. It was somewhat less ap- 
parent, all this while, that he also epitomized the goal of every Ameri- 
can novelist, which was to render his own convictions, however sober- 
ing, and to be popular at the same time. 

Between his first book in 1899 and his death in 1946 Tarkington 
published thirty-nine volumes of prose fiction, and wrote twenty-one 
plays for the professional stage. He also wrote six plays for magazine 
and radio audiences, two volumes of essays, and a book about painting. 
A fragmentary novel and three short novels were published after he 

Still, there was more. Six magazine serials remain among his uncol- 
lected writings, along with two hundred short stories and as many 
articles and editorial essays. But he had more than facility and stam- 
ina, he had the knack of satisfying the magazines. Every novel and 
story Tarkington published in book form all thirty-nine volumes- 
first appeared in a magazine. 

The means of this accomplishment bear looking into, for they were 
the means of counterfeiting the dream-world. To begin with, Tarking- 
ton observed the technical necessities of serial fiction. His novels are 
simple constructions in the narrative present with seldom a detour into 
the past. The main conflict is evident in the first episode, and smaller 
conflicts which amplify it along the way are comfortably divisible into 
installments. The characters all appear early; they are identified by their 
responses to the situation in which they are discovered (as in a stage 
play), and except for the hero or the heroine they remain unchanged. 
There is nothing to confuse the suspense between installments. 

Tarkington took particular care of his conflict by building his novels 
backward or so he said. (Any author's statement about "how he did 

ioo The Denatured Novel 

it" is always questionable. ) Tarkington once described his method in 
a publicity questionnaire which the Literary Guild customarily sends 
out to its authors, although the Guild never publicized this particular 
discovery. "I make a detailed outline or synopsis before beginning the 
manuscript of the novel," Tarkington wrote. "This synopsis begins at 
the end or final scene and goes backward from the opening of the story. 
I use this method because it produces its own order of action." And 
this, he explained, "means that you cannot change the course of the 
story without producing disorder, fumbling and uncertainty." 

Unlike the serialist, Temple Bailey, who relied on sharing the desires 
which her readers wrote to her as she continued her stories, Tarking- 
ton would not change a story's resolution. He had to plan for it to turn 
out all right. Despite its dubious function of publicity, Tarkington's 
statement happens to explain plausibly what went wrong with so many 
of his novels. 

One such failure is The Lorenzo Bunch (1936), about neighboring 
families in an apartment house. The novel was supposed to feature a 
conflict between a husband and wife, each with a liaison on the other 
side of town. Early in the story, however, the neighbors in this apart- 
ment house begin to come alive and to steal the scene from the hero 
and heroine. These limited people are so bored by their confined ex- 
istence that they have resigned their private lives to a kind of com- 
munal emotion. Tarkington had something big going here when it 
came time to resolve the crisis. The synopsis had called for the heroine's 
panting escape from an extra-marital situation, but meanwhile too 
many live characters had got involved for any quick solution to be 
convincing, and Tarkington needed more room than five installments 
of McCalVs could accommodate. 

Firm, pre-meditated resolutions often jostled Tarkington's novels 
out of proportion by denying any real development of the characters 
or any resonance to their involvements. The problem was that in order 
to express his own attitudes about human shortcomings Tarkington 
had to tell two stories at once his own and the magazine's. Turmoil 
(1915), for instance, is about a father trespassing on his sons in order 
to build an industrial empire for himself, but this gets absorbed into a 
magazine story about one of the sons. This is typical. 

Slick Nirvana 101 

This magazine story is always the same. It is the story of Parsifal or 
of Jack-and-the-Beanstalk: of the lowly hero who is scorned but who 
proves himself and then collects the rewards of his courage and virtue. 
It is the exile and return of Monsieur Beaucaire, of the Gentleman 
from Indiana, of Bibbs in Turmoil and Dan Oliphant in The Mid- 
lander, (1923), of Tom Vanrevel and Hatcher Ide and Irvie Pease. 
What each of these Tarkington heroes most resembles is all of the 

This was Tarkington's permanent sub-story. He was skillful at local- 
izing it and merging it with recognizable conflicts in the American 
society. In its purest form, uncontaminated by any ideas, Tarkington 
called it The Conquest of Canaan (1905), a fable that explains all of 
his fiction % 

Canaan is a small, conventional Midwestern community in the 
i88o's, whose 1 citizens measure themselves by money and "position," 
and who have no room in their hearts for the hero. On one side of 
town at the end of a long, comfortable lawn, in a Victorian mansion 
lives Judge Martin Pike, a large man who has always "stood for all 
that was respectable and financial/' On the other side of town, along 
the river front and near a disreputable dance-hall arej called Beaver 
Beach young Joe Louden has grown up. 

Louden is a lawyer and an outcast in his own town except for the 
loyal friendship of Ariel Tabor. She has grown up with Joe, and was 
something of an outcast herself until her uncle's legacy and a trip to 
Paris bestowed beauty and desirability upon her. When it develops 
that her trustee, Judge Pikq, has embezzled her inheritance, it becomes 
Joe Louden's formidable chore to rescue what is left of the girl's money. 
One other obstacle is the presence of Joe's urbane step-brother who 
went East to college and has now returned to ingratiate himself with 
Judge Pike. 

Here ended the first installment in Harpers Magazine, but the 
reader who wanted to jump at conclusions had to wait for five more 
installments until Judge Pike was defeated (exposed, in fact, as the 
owner of the disreputable Beaver Beach); and it took one more install- 
ment after that for Joe Louden to head off another suitor and win 
Ariel Tabor for his own. 

102 The Denatured Novel 

The heroine of The Conquest is one of Tarkington's permanent fix- 
tures. She stands for health and vitality, tolerance and understanding, 
and sensitivity. Except for one or two tiny but obvious imperfections 
she is an abstraction of all positive human qualities; being abstract, she 
satisfies the need for what the trade calls "reader-identity." 

Outside the story this heroine is no more credible than the picture 
of the girl on the magazine cover, but within it she fulfills a necessary 
function. The hero is at odds with his world, and someone must under- 
stand him and champion whatever cause he may have. Since he is not 
perfect, yet perfectible, someone must correct his mistakes and fur- 
nish him with an ideal. Since he must arrive at a marriageable age 
anyway, this virtuous woman's love denotes his earthly as well as his 
spiritual success. She is not a person but a project. 

Tarkington later invented a childhood for this typical heroine in 
Kate Fennigate ( 1943) . After burying her alcoholic father and her in- 
valid mother this orphan girl grows up in a household where everyone 
uses her good intentions and selfishly exploits her. But this strenuous 
training in virtue pays off, for after Kate has grown up and married she 
discovers that her acquired patience is really her best asset in subtly 
managing her husband's life. 

The heroine is just one of the caricatures in Canaan. Another is the 
hero's step-brother, who stands for everything the hero is not. This 
counterpart turns up in most of Tarkington's novels, and like Cinder- 
ella's older sisters, he always offers the alternative. Over the years Tar- 
kington worked every possible permutation of the counterpart char- 
acters, with heroes and heroines, pleasant and unpleasant. 

The counterpart is a technical device with distinct advantages for 
the slick story. It is quick, it increases the possibility of explicit con- 
flict; and the reader is assured of the virtue of one character by indulg- 
ing in the unpleasantness of the other. It is also a kind of mirror-writing 
in which, however, the object and the image reflect only unreality. It 
is no different in Tarkington's fiction than it was in the novels of 
G. P. R. James, one hundred years ago; and E. P. Whipple's judgment 
still stands: "We demand human beings, not embodied antitheses, or 
personified qualities, thoughts, or passions." 

Tarkington contrived a certain efficiency with this device by having 

Slick Nirvana 103 

his counterparts take sides in the conflict of some larger ideas, which 
is how he managed to dramatize his own opinions. He also occasionally 
exploited the possibilities of the character with the split personality. 
Usually, this double-character has temporarily acquired a new appear- 
ance: by an accident (in The Guest of Quesnay, 1908) or by education 
(in Mirthful Haven, 1930), or by being spoiled (in Image of Josephine, 
1945) . The new state of affairs is always dreadful until the exiled char- 
acter begins to discover his old self again. Then he returns to conquer 

Tarkington made this double-character convincing only twice, in two 
different books: a farce called Presenting Lily Mars (1933), and Alice 
Adams ( 1921 ) . Lily Mars has a job in a theatrical company and a small 
part in a play about to open. She is so obsessed with making good that 
she creates an off-stage role for herself which upsets the entire produc- 
tion. Her double-character is convincing enough for the farce, but only 
in Alice Adams does the convention reveal anything about live human 

Overcome by her own mediocrity, Alice Adams manages to convince 
herself that she is somebody else, a Miss Adams with a self-respecting 
father, with money and clothes and a college education, a girl whose 
charm and beauty is the toast of her many friends. But even as she car- 
ries on this public performance she realizes its imminent failure and 
tries to evaluate her two selves. 

Alice's father also devises an escape from the dead-end of his own 
life; and he too deceives himself to anyone who will listen. Having stag- 
nated for years as the head of the "sundries" department in a drug com- 
pany, he dreams of manufacturing a new glue with a formula he long 
ago discovered. His wife is willing to be deluded. Father and daughter 
and mother press farther into their own imaginations, each repeating 
the others' deceptions, each believing in his own. Each comments on 
the others without knowing it, and the analogy of their lives is the only 
dreary truth there is. 

This is Tarkington's finest novel. It is resonant and vital. It is also 
an explicit denial of the dream world. The dream is a fake which 
thrives on fear, which feeds on dignity until there is no dignity left. 
Alice Adams uses the same fixtures as Tarkington's other novels use: 

1 04 The Denatured Novel 

the conventional heroine and the conventional double-character; but 
this time the conventions themselves are the subject and subject to 

Alice Adams was the great irony of Tarkington's career. It was the 
only time he made the dream-world completely real and then, only 
by demolishing it. And it was the only time he ever tried. Although he 
continued to write for twenty-five years after its publication, his later 
novels were merely more skillful counterfeits than their predecessors. 
This fact does not square very well with what he once told Kenneth 
Roberts about the permanence of books and the transience of maga- 

Tarkington wrote, in the Literary Guild's questionnaire: "My books 
are rather various both in substance and in manner. One of them tried 
to do and be one thing, and one of them tried to do and be another." 
But only one of them is really various and wholly lifelike. What he 
meant by 'Variety" can be explained, in the others, by the ingenious 
transfer of old fixtures to new settings. 

The most valid criticism of Tarkington's fiction is that there is too 
much of it. It belongs to that great bulk which Elmer Davis called "not 
what somebody wanted to write, but what somebody else wanted to 
get written." The reason there was too much of it lay in the economics 
of magazine publishing. Booth Tarkington, the magazinist, epitomizes 
the paralyzing success of American authorship, for he counterfeited 
the slick Nirvana so well that he came to believe in it. 

Chapter V 

Hollywood Pay-Off 

THE American motion picture industry, which began in 1896, during 
the past fifty years has become the trade publisher's largest customer. 
Within this brief period the movies have eroded the novel form more 
than anything else in the novel's history. 

The film itself and the economic organization of the industry have 
caused this damage. The film narrative is limiting. It necessarily 
abridges the complexities of any subject it narrates. It is more suscep- 
tible than the novel to clear, uncomplicated attitudes, but American 
producers have nearly always forced just such clear and uncomplicated 
attitudes upon it. The size of the industry extends this damage. 

The film industry outgrew the penny arcades and shooting galleries 
of the iSgo's so rapidly that by 1939 it grossed annually more than 
three-fourths, and netted more than ninety per cent, of the income of 
the entire entertainment field in the United States. By 1925 the film 
studios were paying prices in six figures for the film rights to novels. By 
1931 Cheney's Economic Survey of the Book Industry described a new 
occupational disease called "novelist's nystagmus," which was "caused 
by keeping one eye on the typewriter and the other on Hollywood." 

The same factors which have caused the film industry to give way to 
television in recent years have even increased purchases of fiction by 
motion picture studios. The cost of manufacturing and distributing 
motion pictures rose so high after World War II that most producers 
merely leased the facilities of a few giant corporations already in busi- 
ness. The possibility of producing a movie on credit and without cap- 


1 06 The Denatured Novel 

ital investment has enticed many venturesome people, and some have 
succeeded. There are 125 so-called independent producers now buying 
novels for films, none of whom were even in business fifteen years ago. 

This bidding for literary property has driven prices sky-high, al- 
though the bidding goes highest among the few established corpora- 
tions. At the moment the record is held by zoth Century-Fox, which 
commissioned Grace Metalious to write a sequel to her notorious best 
seller Peyton Place (1956) for $265,000. The record is only temporary, 
however, because high prices have great publicity value. 

This was an important transaction for more than just the money in- 
volved, because the whole idea of this second novel was instigated not 
by the author or even the publisher or the reprinter, but by the film 
company. The amount of money had nothing to do with the sequel 
itself which did not even exist at the time except as an idea in the 
producer's mind. The price was established solely on the strength of 
the publicity which had been generated for Peyton Place, and which 
had helped to sell eight million copies of that novel in all editions and 
to promote a successful motion picture. Some publicity, it was thought, 
would doubtless rub off on another novel by the same author, and 
another picture. So the studio got enough copy for the new scenario, 
in the approved Hollywood fashion by commissioning the author to 
write a new manuscript to be called Return to Peyton Place. 

Although the price was higher than usual, this whole transaction was 
characteristic. Publishers' Weekly was talking about "cinema novels" 
back in 1917. Today, although the film rights to most novels fetch less 
than $35,000, Hollywood buys literary property but really pays for its 
notoriety. The merest possibility of notoriety keeps the film companies 
always interested in an editorial partnership with authors. 

Suppose an author has a manuscript in progress and a contract with 
a publisher. If a film company were to buy an option on the completed 
work, all parties would benefit, including the agent who handled it. 
The film company would get a title and an author's name. The author 
would have more money to start with, shared equally with the pub- 
lisher. Everyone would have a surer piece of merchandise, and if the 
book were finally filmed, the publisher and the producer would profit 
by promoting the book together. 

Hollywood Pay-Oft 107 

The film company is not interested solely in an easily adaptable 
novel, it wants the best novel it can get. This is not a matter of being 
noble; it is simply good business. When the time conies, a screen 
writer can translate the book into a film better than its author can. 
Meanwhile, any book a lot of people will talk about is the property the 
producer wants most. The motion picture studio will be glad to retain, 
of course, whatever movie values the novel itself may have. 

This enlightened attitude prevails only because all other means of 
editorial collaboration have usually failed, although there have been a 
great many experiments, all of them undertaken in the film industry's 
urgent need for story material. Ever since 1900, when the Frenchman, 
Georges Melies, filmed the first multi-scene narrative (the sequence of 
Cinderella), the industry has needed more stories than it could get. 
The motion picture studios have scavenged fiction from the public do- 
main and purchased what they had to, even as they hired gangs of 
writers to generate more. 

Published fiction had a particular appeal to the movie industry long 
before the idea of coordinated book promotion. In the early days, under 
threat of organized boycott for filming censorable stories, producers 
found a reasonable guarantee of respectability in the fact of a novel's 
prior publication. If a story were questionable, at least the producer 
could take the position that it was not his fault. If a book had some 
vintage, so much the better. Thus, by 191 3 there had already been four 
motion pictures based on the Leather Stocking Tales and three on 
The Scarlet Letter. 

Adapting published fiction, however, quickly involved the problem 
of copyrights. A test case of the film studios 7 liability for the use of 
literary property, involving finally a Supreme Court decision, changed 
the whole prospect of trade-book publishing. In 1907 Sidney Olcott 
produced for Kalem studio a motion picture of Lew Wallace's Ben-Hur 
(1880), and like the novel the film found an immense audience. Still 
one of the most expensive films ever made, it cost an estimated $6,000,- 
ooo. Partly because of its notoriety, Wallace's publisher brought suit. 

The case was settled in 1911, with a judgment against Kalem for 
$50,000. During the litigation other studios began paying token sums 
for literary property: D. W. Griffith, for instance, paid $100 for the 

io8 The Denatured Novel 

film rights to Helen Hunt Jackson's novel, Ramona (1884). But the 
decision in the Ben-Hur case sanctioned new prices for fiction. Since 
the industry now had to purchase fiction, a studio was wise to buy a 
novel with proven salability. 

Best-selling fiction consequently became valuable property. In 1920 
Eleanor Porter sold the film rights to Pollyanna (1913) for $80,000; 
current novels by Rex Beach, Jack London, Ben Ames Williams, Booth 
Tarkington, Richard Harding Davis and Irving Bacheller brought com- 
parable sums for screenplays. The novel's most valuable asset, of 
course, was its author's name. This fact led the studios to experiment 
in editorial collaboration. They began to hire novelists to make their 
own adaptations. 

Occasionally it worked. The film industry's most efficient novelist 
was Zane Grey, who finally adopted the system of sending the manu- 
script chapter of a novel to the film studio even before he sent it to 
his publisher. One result of this efficiency was that Grey sold the film 
rights to more novels for something over forty motion pictures than 
any other writer has ever sold. Samuel Goldwyn succeeded for awhile 
in hiring novelists to make their own adaptations. In 1919 he organized, 
within the Goldwyn Company, Eminent Authors Pictures Incorpo- 
rated, a group of writers who would direct the screen adaptation of 
their own fiction: Gertrude Atherton, Mary Roberts Rinehart, Rupert 
Hughes, Gouverneur Morris, Basil King and Rex Beach. This organi- 
zation led the way to all kinds of contracts between authors and 
studios, involving publishing rights to publish present and future works 
and the author's services for original screenplays. 

During the 1920'$ the writer became the highest-paid worker in the 
movie industry. Whereas he had once earned $5 to $15 for a story idea, 
in 1906, he could now earn $1,000 to $2,500 a week in salary or $10,000 
to $25,000, for piecework. Meanwhile the studio profited by his name. 
But most of the contracts worked only in theory. Hiring well-known 
writers continued through the thirties, but authors and producers alike 
began to discover diminishing returns. 

The trouble was that the studios bought properties and services they 
could never use. It is customary in the film industry, unlike the book 
trade, for a writer to submit his work simultaneously to several pro- 

Hollywood Pay-Oft ' 109 

ducers. Buying competitively, the studios not only overstocked their 
manuscripts but made most of their purchases sight unseen a routine 
procedure which would horrify any publisher. Like the purchased 
novel, the purchased author's services seldom fitted the requirements. 
Hiring novelists to write scenarios turned out to be a misappropriation 
of skills, rather like hiring a cabinet maker for carpentry. 

Scores of writers have fled from Hollywood and lived to tell about it 
in sour novels or in angry statements which the magazines are more 
than happy to publish. Writers have always complained even Sam 
Goldwyn's Eminent Authors that they were hired not to write but 
to practice some fancied verbal specialty, such as plots or gags or dia- 
logues, and to collaborate with other writers who were also paid not to 
write. The waste of effort and money over stories that were abandoned 
was infuriating, but their bitterest complaint was over the ruining of 
literary property. After the studio improved what he had written the 
author could scarcely recognize the mess. 

Adapting a story to a screenplay in Hollywood is a matter of shrewd 
improvisation. Writers are paid to reassemble ideas borrowed from 
screenplays that have already succeeded. This is exactly what has al- 
ways happened in the book trade. And when the author and screen- 
writer, Mildred Cram, described this process in the American Spec- 
tator Year Book (1934), she spoke for all embattled writers in Holly- 
wood. "You aje, let us say, writing a scene of seduction. Someone re- 
minds you of the great seduction scene in 'The Guardsman/ Formula 
Number Eighty. Or someone else calls your attention to the sure-fire 
seduction in 'Red Dust.' " 

They like 'em strong, fundamental, earthy. The anemic heroine is 
out. Let's write this straight from the shoulder maybe with a laugh 
you know, like that scene in "Farewell to Arms," where she laughed at 
herself. That was a wow. People loved it! 

Presto, your seduction scene is a patchwork of the best seduction 
scenes! Formulas eighty to nine hundred, inclusive! 

Every writer in Hollywood sooner or later discovers that what is "ac- 
ceptable" in fiction has already been prescribed for him. The only tol- 

1 10 The Denatured Novel 

erable invention what he is hired for is some new twist of a few old 
and salable ideas. This is so because the screenplay is not an autono- 
mous art form, and never has been; it has always been a by-product of 
something else. In the whole history of the industry the motion picture 
itself has served one master after another. 

Right from the start the film industry has belonged to the manufac- 
turers of production equipment (cameras and projectors, and later, 
sound recording and projecting devices). The manufacturer has re- 
garded the tnotion picture as merely an adjunct to his main line of busi- 
ness. This emphasis began with the industry in 1896, in New York City, 
in Koster and Rial's Music Hall by utter chance on Shakespeare's 
birthday when a machine known as the Vitascope projected onto a 
screen a filmstrip lasting about a minute and a half. 

The projector was Thomas Armat's improvement of the Kinetescope 
which Thomas Edison had invented in 1889, partially patented in 
1891, and installed in penny arcades. Edison's peep-show had allowed 
only one viewer at a time to see the filmstrip, but Armat's projector, 
enabling an audience, made real profit possible. For seven or eight 
years, Armat, Edison, and several other manufacturers experimented 
with cameras and with the production of motion pictures, and by 1903 
they had established a pattern of control by leasing to exhibitors both 
their projectors and their films. They maintained this control. 

In 1908 four manufacturers including Edison and Armat pooled 
their patents and organized the Motion Picture Patents Company, to 
restrain competition and to keep the market for themselves. Although 
this trust was liquidated in the courts in 1917, three years after it had 
lost most of the market to independent manufacturers, it had forced a 
characteristic growth on the industry. Controlling the market for mo- 
tion pictures mean controlling three separate but related operations: 
production, distribution, and exhibition; and in order to fight the trust, 
independent producers had to achieve similar control. 

The economic development of the industry after 1920 merely con- 
centrated control in fewer and larger corporations, each combining pro- 
duction and distribution facilities with a chain of theaters. This was 
speeded up after the invention of the talking picture in 1926. By 1929 
the silent film was no longer profitable, and the motion picture com- 

Holly wood Pay-Oft - 111 

panics and their film studios gradually fell under the dominion of 
the manufacturers of recording as well as projecting equipment: West- 
ern Electric, a subsidiary of the Bell system, and later, R.C.A., which 
bought a film studio and a chain of theaters in order to market its own 

By 1929 there were five major combinations engaged in the produc- 
tion, distribution, and exhibition of films: Warner Brothers, Para- 
mount, Loew's, zoth Century-Fox, and Radio-Keith-Orpheum. Within 
each of these combinations film production was determined not by the 
studios but by distributors and exhibitors. After the stock-market crash 
these major combinations could no longer finance new operations out 
of earnings, and outside capital (from banks, real estate holders, and 
motion picture finance companies) exerted new influence on the pro- 
duction of films. 

Economic Control of the Motion Picture Industry (1944), by Mae 
D. Huettig, gets to the significance of this within its first chapter: "The 
fact that the major producers of film have not been forced to rely ex- 
clusively on the excellence of the product itself for profit may contrib- 
ute to an understanding of many questions concerning the progress of 
the film as an art form." Every change in the industry has resulted in 
making this art form a more efficient source of income for capital in- 
vested in related properties. 

This includes literary properties. Since Armat's invention of the 
Vitascope, every economic crisis in the industry has increased its de- 
mand on the book trade for more fiction. Competition between inde- 
pendent producers and the Motion Picture Patents Company, after 
1909, resulted in the feature-length film. Two world wars shut off Euro- 
pean competition even as they increased the world market for Ameri- 
can motion pictures. The invention of the talkies added major pro- 
ducing companies to the industry, and caused an increase in the pro- 
duction of films. The box-office decline in the early i93o 7 s prompted 
the double-feature program, offering twice as many reels for the view- 
er's money. Every major change in the business increased the need for 
screenplays and the value of literary property. 

After World War II independent producers joined the industry in 
force. Their presence expresses more than a growing need of story ma- 

112 The Denatured Novel 

terial. They typify Hollywood's precarious financial position. They are 
independent in name only, since their business is entirely answerable 
to the banks and the distributing companies who control their credit. 
The so-called independent sets out to produce a film with nothing but 
a few contracts and a large debt. His creditors have not loaned him a 
clime without being satisfied that his product will make money. 

The values of fiction arc box-office values, as far as the screenplay 
is concerned. The fact that businessmen have always controlled the 
major companies which the screenplay serves, accounts for some of the 
weird ideas the so-called "business values" which box-office sales are 
supposed to reflect. One axiom is that the film serves one vast public 
which it must please at all costs. Another axiom, until the independent 
producers (artists mostly) proved otherwise, was that the more a prod- 
uct costs the better it is. 

These two axioms have generally limited the content of motion pic- 
tures to more expensive elaborations of what has already been pre- 
sumed to please everyone. The industry's rising costs, which increase 
the capital risk, have always made it prudent to continue this policy of 
elaborate imitation. A major producer-distributing company with its 
staggering overhead must spend at least $1,000,000 for every one of its 
fifteen or twenty pictures a year. Merely breaking even is too great a 

The so-called "business values" partly explain the manipulation of 
literary property so that it will all look alike on the screen, but there is 
another economic cause which insures manipulation. This is the star 
system, which has hypnotized the entire industry: the process of pub- 
licizing actors and then building movies to exploit their box-office ap- 
peal. The star system functioned as early as 1910. It was one of the 
weapons used by independent producers in their battle with the li- 
censed members of the Motion Picture Patents Company, 

Carl Laemmle began the star system and Adolf Zukor amplified it, 
stressing certain personalities rather than motion pictures. But by hav- 
ing featured personalities on the pay rolls, producers then had to pro- 
tect their investments with large salaries. Box-office response to the 
first of these stars, Florence Lawrence and King Baggott, then Mary 
Pickford, Charlie Chaplin, John Barrymore, Minnie Maddern Fiske 

Hollywood Pay-Oft ' 113 

and James K. Hackett, justified their inflated salaries. Justification, in 
each case, meant more pictures like the last. Audiences typed the stars 
good or wicked, strong, tender, seductive, wistful or whatever and 
these traits had to be repeated. 

Type casting resulted in reassembling over and over the same limited 
and calculable stories to fit a star's public personality. The star system 
even influenced the distribution of motion pictures. When theater 
owners demanded more of the stars, the producer used this bargaining 
power to lease his other, less publicized and less costly pictures. Adolf 
Zukor's producing company, Famous Players, classified its releases as 
"A" "B" or "C" pictures, depending on cost and an estimation of box- 
office appeal, a rating system which became indispensable to the in- 
dustry after the introduction of double-features in the 1930'$. 

The star system and the classification of first- and second-class mer- 
chandise affected the purchase of literary property in a particular man- 
ner. Any scenario begins from one of two sources: an "original," writ- 
ten to prescription; or a published work of fiction, drama, biography, or 
history. Originals are characteristically the source of "B" pictures, 
which include nearly all of the Westerns, but most "A" pictures begin 
with a book already published. 

By Hollywood's Alice-in-Wonderland logic, anything that costs 
more is better. The star needs a vehicle whose notoriety will match his 
own. If the author's name has news value, so much the better. The 
announcement of an inflated sales price appears to endorse the worth 
of the literary property, but its worth is only its ability to attract atten- 
ion. What it becomes in the film is something else again. 

The look-alike quality of the movies is never left to chance; it is a 
studied quality. The absolute values are the same as those in the dream- 
world of the slick magazines. The pursuit of happiness engages the 
hero and the heroine, and these tabloid characters are revealed only 
with regard to the particular conflict which besets their pursuit. In the 
resolution of this conflict everyone is paid off in kind. All that varies 
(with the same regularity) is how the pay-off is delivered, how the 
values of the hero and the heroine are vindicated. 

The obvious fixtures of this dream-world have scarcely changed since 
they were cataloged under the auspices of the Motion Picture Research 

H4 L ne uenaiurea i\ o vei 

Council twenty-five years ago. Edgar Dale's comprehensive report, The 
Content of Motion Pictures (1935), is based on the evidence, discov- 
ered by a team of researchers in 1,500 motion pictures released between 
1920 and 1932. This is what they found. 

Romantic love and crime were the subjects of most of these motion 
pictures. There were far more characters living on high income in the 
films than in actuality, and material well-being was what heroes and 
heroines wanted. As for the superficial realism of setting, the urban 
scene, particularly New York City, predominated. The most frequently 
used interior set was the bedroom, although love-making, symbolized 
by the kiss and embrace, occurred most commonly in the living room. 
Illicit love was the intent of the villains more than the heroes. (Movie 
villains, characteristically in their thirties, are older and more faded 
than heroes.) The happy ending, enforced by the star system, clearly 

This inventory gains significance, not originally intended, in view of 
the palling sameness of the substance. Even before the Dale report the 
expedience of the movies aroused indignation. At the height of the in- 
dustry's independence of outside and cautious capital, when movie 
companies cojjld produce whatever they wanted, John Gould Fletcher 
indicted producing companies for what he called The Crisis of the 
Film (1929). 

"To produce something that could be understood quickly, sold 
quickly, repeated quickly, has been mainly their aim," Fletcher 
charged. "Their aim was to give the public exactly what it wanted, and 
if they found out that the public wanted one type of thing, to go on 
giving that thing with as little variation as possible." This resulted, he 
said, in "propaganda for the emotional monotony, the naive morality, 
the sham luxury . . . the sentimental, and the acrobatic that are so com- 
mon in the United States." 

Only the costumes and the customs have changed with the times. 
The experience remains. The movies kept pace with the change in 
sexual mores, for instance, from the Mary Pickford vehicle, Pollyanna 
(1920), through the vamps played by Theda Bara, to the aggressive, 
likable, worldly woman personified by Jean Harlow, Carole Lombard, 
and Mae West. 

Hollywood Pay-Off '115 

The new films kept up with the fashions by mocking the mores of 
their predecessors. The vamp pictures mocked the behavior of the 
Pollyannas, and the Mae West pictures mocked the prudery with 
which the vamp had been characteristically defeated in her desires. But 
all of these films shared the same kind of moral posture. All of them 
reveled in the behavior assumed at that moment to be a trespass, before 
retributing it. 

Such posturing, of course, belongs to popular entertainment. The 
movies have not tampered with vulgar expedience. They have simply 
exploited the same appeal with which Samuel Richardson's Pamela 
made its prurient plea for chastity. That book, like its heroine, sold 
extraordinarily well. The movies have merely rearranged the pay-off. 
The story of the heroine breathless from the lover's pursuit but holding 
out for marriage has been filmed under one title or another since the 
movie industry began. 

Moral posturing encompasses more than chastity, as the crime pic- 
tures demonstrated after the Legion of Decency was founded in 1934. 
Prior to this censoring organization films presented crime without edi- 
torial comment, more or less in the guise of documenting American 
urban life. When the spokesmen for censoring groups had had enough 
of this, however, crime pictures brought the criminal to harsh justice 
in the last few minutes of the film, having already celebrated for an 
hour-and-a-quarter the criminal's modus operandi. Sham moralizing 
has also marked most of the so-called sociological films about the rights 
of labor, the problems of White and Negro, divorce dilemmas, juve- 
nile delinquency, and the ethics of Big Business. 

Pleasing the public has been carefully assured in the film industry. 
Courting pleasure or avoiding trespass takes the form of pre-censor- 
ship, a review conducted before a picture's release and designed to 
avoid costly litigation. Pre-censorship has existed since 1909, when the 
Motion Picture Patents Company responded to the fears of church- 
men and public officials (which newspapers and magazines gleefully 
printed) that the movies were conducting an organized public de- 
moralization. The Company organized a censorship board to review 
the productions of its member studios. 

With this precedent the major film companies adopted in 1930 a list 

1 1 6 The Denatured Novel 

of commandments and prohibitions known as the Motion Picture 
Production Code. The Code is still enforced by a board of the indus- 
try's trade association once known as the Hayes Office nominally 
independent but actually a sounding board for private, vocal lobbies 
political, religious, professional and civic, and comprised of the ubiqui- 
tous club woman engaged in special pleading and in the state of 
public morals. The power of this official censoring group lies in its 
right to withhold its endorsement of any motion picture. On the basis 
of this awesome power it "recommends" the deletion of anything of- 
fensive to the Motion Picture Code. 

The Code is notorious for its prohibition of the details of sexual 
intimacy in or out of marriage, whether of fruition or frustration in 
fact, its taboo of all explicit references to any of the biological functions 
of animal or human characters. The prohibition also governs thematic 
treatment of these details. Love must end in marriage; divorce is for- 
bidden. Fitting punishment must follow a display of sin or crime. 
There may be repentance in the motion pictures but no redemption. 

Unnumbered writers novelists, lawyers, historians, clerics and poets 
have protested that this "tattered philosophy of free will" contradicts 
Christian ethics and the moral values the Code attempts to preserve; 
but to no avail. Two men peculiarly suited to the task once attempted 
to fix the responsibility for film censorship: Morris Ernst, this nation's 
most prominent attorney in censorship litigation after the Ulysses 
court case; and Pare Lorentz, producer and director of epochal docu- 
mentary films. Their scrutiny, recorded in Censored: the Private Life of 
the Movies (1930), led them time and again to "the gross, vapid illu- 
sions of life contained in the ordinary movie." 

Ernst arid Lorentz explained the operation of the Hayes office, and 
discovered personal reasons for militant bigotry among dozens of indi- 
vidual censors; but somehow the marriage of bigotry and expedience 
was larger than the sum of discoverable details. Responsibility was 
everywhere. They concluded: "The masses made the movie. It is a 
nationalism; a sectionalism of thought, desire, frustration, written 

Exploiting this desire and frustration is the business of the movies. 
Hollywood: The Dream Factory (1950) is what Hortense Powder- 

Hollywood Pay-Off 117 

maker called her anthropological study of it, and the sum of all its 
expedience she called a taboo against mankind. "No one goes to bat for 
the human species; no one seems to care that mankind is presented 
falsely, that the majority of movie portrayals are untrue." In all the 
special pleas for doctors or Mexicans or ministers or Rotarians or 
Negroes, "no worthy organization protests that human beings are 
shown as passive, unfeeling robots." In a chapter called "Taboos" she 

Man, according to Hollywood, is either completely good, or bad. His 
personality is static, rarely showing any development either in growth 
or regression. The villain is a blackened sinner who can do no good and 
who cannot be saved; while the hero is a glamorous being, who can do 
no wrong of his own volition, and who is always rewarded. 

The movies manufacture stereotypes disguised as persons. They 
rarely bother about complex human beings. This is a result of the 
manufacturer's belief that he must gratify the consumers' wishes in 
order to stay in business. But making profit entails more of a compro- 
mise than merely expressing certain acceptable attitudes. To visualize 
these attitudes on film means mutilating the literature from which they 
are abstracted. There is no way around it. 

The film form always emphasizes the prominent features of any 
fiction. It simplifies the novel by caricaturing it. For the novel's system 
of analogies the film substitutes an immediacy which the novel does 
not have the reader becomes a viewer. But the film cannot accommo- 
date the novel's complexity. When the producer uses the film to visual- 
ize these unrealistic attitudes, therefore, he merely aggravates the film's 
natural distortion of the novel. 

Adapting a novel to the film means more than just converting it. The 
first chapter of George Bluestone's Novels into Film ( 1957) concludes 
on this. What the "filmist" adapts "is a kind of paraphrase of the 

He looks not to the organic novel, whose language is inseparable from 
its theme, but to characters and incidents which have somehow de- 

1 1 8 The Denatured Novel 

tached themselves from language and, like the heroes of folk legends, 
have achieved a mythic life of their own. 

One novelist has made a convincing case of this reasoning by turn 
ing the usual process around: by building a novel out of a screenplay 
and then explaining why. Budd Schulberg wrote the scenario for 
Waterfront (1955) about racketeering among the longshoremen of 
the Port of New York. Then, dissatisfied with the filmed result, he 
wrote a novel based on the same story. 

This film had every advantage of an intelligent production. The 
author knew his subject and had something to say; he was allowed to 
control the scenario; and the direction and the acting amplified his 
story with sensitive understanding. It was a superb film and justly won 
several Motion Picture Academy awards. But the film was not enough; 
it told the truth of the lawless world of the waterfront, but not the 
whole truth. Schulberg wanted to represent the dilemma of many 
responsibilities for this lawlessness. 

Schulberg's real subject was not the waterfront activity but the 
awareness of it, which he described as "the deeper truth of inconclu- 
siveness." His article, "Why Write It When You Can't Sell It to the 
Pictures," in Saturday Review (September 3, 1955), explains why the 
scenario could not accommodate his subject. The film, he said, has a 
relentless force. It has its own tight logic which denies the ranging 
kind of aesthetic experiences: "A film must act, a book has time to 
think and wonder." 

The film is an art of high points The novel is an art of high, middle, 

and low points The film does best when it concentrates on a single 

character. ... It tends to lose itself in the ramifications. ... It has no 
time for what I call the essential digressions. 

By "essential" digressions Schulberg meant the exploration of compli- 
cated, contradictory characters and of their backgrounds, which only a 
novel can afford. 

Hollywood Pay-Off 1 19 

The film must go from significant episode to more significant episode 
in a constantly mounting pattern. ... It cannot wander as life wanders, 
or pause as life always pauses to contemplate the incidental or the un- 

The scenario visualizes a situation. The novel comprehends it. The 
scenario abstracts the motion from that situation for its own subject, 
because motion is the distinguishing property of the film. Everything 
that a movie has to reveal depends on the uninterrupted movement of 
its story. 

This property of continuous motion determines the shape of the 
story. Like the film on its winding spool the action must progress. Its 
continuity derives from successive encounters which build some large 
conflict whose resolution will end the film. Each encounter comprises 
a scene, and the average feature film has time to develop only about 
thirty scenes. Since each scene must advance the main conflict, there 
is no room in the narrative, no space in the film, to explore any periph- 
eral subject. There is no room for what Budd Schulberg calls the deeper 
truth of inconclusiveness. 

Abstracting motion from a novel means selecting and emphasizing a 
conflict. This is standard procedure in adaptation. On the subject of 
"Scenarios," Anita Loos's and John Emerson's Breaking into the 
Movies (1921) still speaks for scores of other screenplay manuals. "In 
choosing your story be sure it has the dramatic quality. It must not 
be rambling; and it must have an element of conflict between opposing 
factors. . . ." This manual describes the kind of conflict that will work: 
between "a man and a woman, a woman and her Destiny, or simply 
Good and Evil which leads up .to a crisis in which the matter is fought 
out and finally settled." 

No mistake about this conflict. The adversaries are obvious, and they 
fight it out until they settle the matter. This recipe for a screenplay has 
no room for the simultaneous analogies which develop and appraise 
the subject. The screenplay reduces the novel to a single conflict which 
it then renders into a sequence of visible scenes. 

The deliberate stages of adaptation, like reduction gears in a ma- 
chine, direct the energy to another plane for more efficient use. At the 

1 2O The Denatured Novel 

first stage the screen writer abstracts from the novel a synopsis of the 
conflict. He then writes a present-tense narrative based on this synopsis, 
which develops the traits and the motives of the characters with ref- 
rence to the conflict. This is known as the "treatment," which is 
enough to make almost any novelist shudder. 

The next draft of the screenplay manuscript is the "continuity," 
which translates the "treatment" into the conventions of film repre- 
sentation. This is the rendering of the novel. The more dramatic 
material is turned into scenes, encounters between the adversaries that 
will build the conflict. The rest of the novel is abandoned. Finally the 
scenario or "shooting-script" refines the "continuity," dividing the 
scenes into camera shots and providing transitions from one scene to 
another. The sequence and variation of these camera shots will ac- 
commodate the need to reveal one fact at a time, while creating sus- 
pense over facts yet to come. 

But this tidy description merely theorizes. Since the organization of 
a film studio makes adaptation a collaborative affair, involving the 
writer with his unit producer, the supervising producer, usually other 
writers and then the director, many versions of each stage of the screen- 
play manuscript modify the idea with which the writer began. Even 
the finished scenario is not a public document but merely a set of in- 
structions for the director, who must often make impromptu changes 
as he shoots the film. 

This whole collaborative business diminishes the writer's jurisdiction 
over the story. What really cancels it is the editing, the process of or- 
ganizing thousands of feet of exposed film into a narrative approxi- 
mating the "continuity." Editing, or "cutting," literally establishes the 
point of yiew of the story by selecting from among the camera shots. 
It creates the film's presiding intelligence. 

Editing can eliminate scenes, change their emphasis, or build new 
scenes. It can change the entire story. Because it controls all of the 
story's proportions, this process takes the place of the cohering pres- 
ence which the author would furnish in a novel. 

The process of "cutting" or editing is nearly as old as the narrative 
film. Two years after Georges Melies filmed Cinderella with a station- 
ary camera, as though it were a stage play, Edwin S. Porter's The Life 

Holly wood Pay-Oft 1 2 1 

of an American Fireman ( 1902 ) demonstrated the vitality of "cutting" 
from one view of a continuous situation to another. By "cutting" Porter 
could show the relevance of separate but simultaneous action, such as 
the burning building in one place and the firemen racing toward it. 
And by "cross-cutting," or showing alternate shots, he could create 

Porter's achievement was a triumph of production. He assembled a 
narrative exclusively in the "cutting-room" from film footage accumu- 
lated in other pictures. His achievement showed the advantages of 
"cutting" for dramatic significance. Later D. W. Griffith, the industry's 
greatest innovator, exploited this technique. One of his versions of the 
"cross-cut," the venerable convention of the chase, has been a part of 
nearly every feature film since Griffith's The Birth of a Nation ( 1914) . 

"Cutting" stresses the immediacy of conflict and breeds suspense 
over the outcome. At every stage it can surprise the viewer. It is a con- 
vention that exploits all the tabloid advantages of the film, and dis- 
tinguishes the film from all other narrative forms. This convention is 
so dominating, in fact, that screenplays are normally written to ac- 
commodate it. 

The scenario capitalizes on the film's capacity for suddenly estab- 
lishing a new point of view. Suspense can be heightened by the sudden 
revelation of new facts, simply by "cutting" to another camera shot or 
another scene. The story itself, therefore, must exploit the unexpected 
turn of action, the sudden reversal, or an ironic twist. The device is 
known in the trade as the "switcheroo" or the "gimmick" or the "top- 
per." If its surprise is funny the device is a "yak"; if sentimental or 
pathetic, it is a "bleeder." 

The sudden twist of events pays off in shock value. The "gimmick" 
is as old as story-telling despite its new slang, but the film's natural 
facility for changing the point of view emphasizes the device of sudden 
reversal. This is how the scenario really denies the novel. The abrupt 
revelation becomes a disguise for the developing awareness which a 
novel can represent but which a film does not have time for. 

The "gimmick" is a sleight-of-hand. The scenarist can plant a fact 
obscurely in the action and later reveal it as cause for the new turn 
of affairs, and if the story is twisted quickly enough the shock will 

122 The Denatured Novel 

deny reasoning. The new turn of affairs may be arbitarary, even un- 
believable, but the scenarist has only to remind the viewer that he has 
seen the evidence before without realizing it. This is quite different, 
however, from the novelist's problem of making the reader believe in 
what happens. 

The "gimmick" or the "switcheroo" is a way of passing off the coun- 
terfeit values of movie fiction those lies which gratify the customers' 
dreams. Because the sudden twist appears to solve a conflict so neatly 
in its own terms, the device apparently justifies those terms: that peo- 
ple are either good or bad and that, good or bad, they are all paid off 
in kind. The "switcheroo" is a dream-device which the scenario ex- 

It would be going too far to say that the scenario and the novel ex- 
clude one another but not much too far. The novel can be adapted to 
the scenario only at the expense of its natural complexity. Sometimes 
this is a good thing, when the novel is confused or too long for what it 
has to say. The movie of James Jones's From Here to Eternity ( 1951 ) 
was far more persuasive than the novel. But this same attenuation 
causes most scenarios to denature the novels they start with. 

The film cannot render conflict without visualizing the adversaries, 
nor can it visualize thought except by action; and at best the "gim- 
mick" is a very stylized action. When a novel emphasizes inaction or 
the futility of action, or when a character's adversary is not another 
character or even a place that can be seen, the film is bound to distort 
it. Yet this is precisely the kind of "unvisual" novel that has developed 
since the end of the nineteenth century. 

By 1900 the doctrine of determinism had begun to influence a few 
American novelists such as Crane, Morris, Dreiser and London; and by 
1920 it was widespread. Toward the end of the nineteenth century the 
economy of Europe and America had seemed to outgrow any individ- 
ual's ability to control or even to cope with it. What seemed an irre- 
vocable loss to the individual needed explaining, and the doctrine of 
"scientific" determinism, as it was called, attempted to explain it. 

This doctrine was based on the layman's enthusiastic misunderstand- 
ing of Darwinian biology and later of the psychology of Freud. It at- 
tempted to explain the individual human being in terms of certain in- 

Hollywood Pay-Off 123 

tangible "forces" presumed to limit or determine him. In the fiction 
which absorbed this irrational doctrine the individual had no efficacy. 
Any action was futile because he could not even define his adversary. 

However irrationally it may have developed, the extraordinary influ- 
ence of "scientific" determinism on contemporary fiction is a fact. Al- 
most every American novel after 1900 in some way acknowledges it. 
This fact tends to make the contemporary novel intractable to any 
abridgment or adaptation. 

When a novel's conflict involves an adversary as shapeless as the 
whole of society or as invisible as the hero's misguided notions of his 
environment, the difference between the novel and its movie is likely 
to be tremendous. When a novel presents an irresolvable situation, 
with the hero suspended in circumstance, then the film narrative 
(which demands the efficacy of action) actually contradicts the novel. 
The well-known filmed versions of two deterministic novels Theo- 
dore Dreiser's An American Tragedy (1925) and Ernest Hemingway's 
To Have and Have Not (1937) show just such distortion. 

The same kind of substitution made the better novel, An American 
Tragedy, into the poorer film. This novel dramatizes the complex cir- 
cumstances by which an individual's environment limits him. An 
American Tragedy tells of a young man who plots the murder of his 
mistress and then executes it in spite of himself. Nearly half the novel 
describes the murder trial, which repeats the convicting evidence so 
many times in so many ways that the book indicts an entire society for 
the youth's crime. Like almost everyone else in this society the hero has 
simply responded to the incentives of prosperity and social privilege. 
The mass of the book is the massive evidence of this implacable situa- 
tion. The novel has no "gimmick"; the pay-off is exactly what the 
cumulating evidence promised it would be. 

Dreiser's novel was first turned into a stage play, a four-act melo- 
drama by Patrick Kearney, which concentrated primarily on the trial 
scene. Then a moving picture based on both the novel and the play 
was produced by Joseph Von Sternberg and released by Paramount in 
1931. Except for the murder and the trial scenes almost no vestige of 
the novel remained. The problem lay in defining the youth's adver- 
sary. The mass of society was not filmable, so the producer imperson- 

1 24 The Denatured Novel 

ated it by law-enforcement officers. But this meant making the youth 
the villain. 

At the outset, therefore, the film reduced the novel to the pursuit 
and trial of a murderer. Dreiser protested against this movie which, he 
said, had turned the novel into a "tabloid murder story." Arguing that 
the right to reproduce the novel did not convey the right to impoverish 
it, the author sought a court injunction to prevent a theater in New 
York City from exhibiting the film. The court's decision against the 
author added a new irony to An American Tragedy. 

The substitution of an obvious and resolvable conflict for the deter- 
minist's view of experience distorted Hemingway's To Have and Have 
Not beyond all recognition. The novel begins with a short story (it is 
really a collection of short stories) about Harry Morgan, fisherman and 
smuggler, whose luck, as his widow says, "went bad first in Cuba. Then 
it kept getting right worse and worse until a Cuban killed him." 

Titled "Spring," "Fall," and "Winter" there is no "Summer" in 
this particular world the stories dramatize the circumstances of Har- 
ry Morgan's losses: his fishing tackle, his arm, his boat and then his 
life. Dying from a bullet wound in his stomach, Morgan mutters his 
own realization of these circumstances: "Ain't got no hasn't got any 
can't really isn't any way out." The author says it took the dying man a 
long time to get it out and all his life to learn it. 

To Have and Have Not is a series of brief glimpses at the lives of 
men and women living around the harbor at Key West. The analogies 
of these scenes suggest that somehow these have-nots all manage to 
help circumstances deprive them of whatever they really want in life. 
They all pathetically cherish some illusion of "having." A few, like 
Harry Morgan and his wife, discover the inadequacy of the illusion, but 
the rest never do. There is no other resolution. 

This book denies the dream. But its author was famous reason 
enough to make a film. There were a few small problems involved, 
however, such as inventing a new conflict with new characters in a new 
setting. The book's first episode, entitled "Spring," contained some 
useful elements; and the scenarists abandoned the rest of the episodes. 
Even so, this was not movie material. In this episode Morgan agrees to 

Holly wood Pay-Off 125 

smuggle twelve Chinamen from Cuba to the United States, then dou- 
ble-crosses his employer by taking the money, murdering him, and land- 
ing the cargo back in Cuba. 

This was scarcely the illusion the movies wanted to sell, so the screen 
writers made some changes. "Spring" became the summer of 1940, and 
the locale was moved from Havana to Martinique, a French possession 
then ruled by the Vichy government. This made Harry Morgan's tres- 
passes acceptable. He could be hired by the French Underground to 
'sabotage the puppet government (which would be represented, of 
course, by a police inspector). It remained only to change Morgan's 
cargo to a member of the Free French forces and to eliminate the 

One other improvement amended an unfortunate omission from the 
original episode. The scenarists wrote into the manuscript the part of 
an American young woman stranded on the island. Profiting by these 
changes, the film could easily represent an acceptable conflict: begin- 
ning with intrigue, developing into a chase, and concluding with a 
"switcheroo" which shifts the balance of power from the police in- 
spector to Morgan. As Anita Loos put it: "A conflict of Good and 
Evil which leads up to a crisis in which the matter is fought out and 
finally settled." 

The scenarists one of whom was William Faulkner showed great 
ingenuity. They twisted Hemingway's story into a permissible conflict 
that could be abruptly resolved. The scenario ignored the book's em- 
phasis on "why" for an easier emphasis on "how." The smuggling of 
human cargo, like the other episodes in the book, implies need to seek 
reason for an apparently unmotivated crime, but the film removes 
precisely this question. With the same change of emphasis the motion 
picture reduced An American Tragedy to a single narrative of pursuit 
and capture, omitting the complexities of why the murder had oc- 

One of the properties of the novel form is the appraisal of conflict, 
and both films effaced this property. But not all novels so stoutly resist 
the "treatment" of the scenarist. If a novel emphasizes only "what" 
and "how" to begin with, and if its conflict already resolves by a 

1 26 The Denatured Novel 

"switcheroo," then making a scenario is no trouble at all. One such 
novel, which also happens to be the story of a murder, is James M. 
Cain's The Postman Always Rings Twice ( 1934) . 

This novel tells how the passions of an itinerant bum and a married 
woman lead them to murder her husband, and how this passion causes 
the death of both murderers. The action turns on several sudden re- 
versals. The first attempt at murder fails fortuitously, when it de- 
velops that the man and woman were already under suspicion. After 
their second, and successful, attempt the clever tactics of their defense 
lawyer frees them. Next they are threatened by a blackmailer whom 
they outwit. Finally, after they are clear of the law and with all evi- 
dence destroyed, an auto accident kills the woman. This accident offers 
evidence enough to convict the bum of her murder, even though there 
was no murder. 

The novel has no theme, no attitude toward its subject. It simply 
tells how the murderers ironically escaped until they were ironically 
punished. Only a few changes were needed for a screenplay that would 
satisfy movie values. The film built up the character of the district at- 
torney, making him the single adversary of the murderers. It simplified 
the trial scenes and eliminated the guilt of insurance companies, which, 
in the novel, profited by letting the murderers go. It added two new 
scenes and new material to two others to make the heroine played by 
Lana Turner more sympathetic, and to give her an added motive for 
murder. It also eliminated some explicit love-making. 

One reason for this faithful adaptation was that the author had al- 
ready written his novel like a movie "continuity." Its sixteen chapters 
are comprised of scenes. Within these scenes, short units sometimes 
only eight or ten lines approximate camera shots. Thus, the book's 
longest description of the wife, at the end of chapter one, efficiently 
offers the only explanation of the mentality of her accomplice. "Ex- 
cept for the shape she really wasn't any raving beauty, but she had a 
sultry look to her, and her lips stuck out in a way that made me want 
to mash them in for her/' 

The book's point of view is a "gimmick." The first-person narrative 
of how the murderers go free turns out to have been written in prison, 
just before the narrator's execution for a crime he did not commit. 

Hollywood Pay-Off 1 27 

Presto! The story has been morally sound all along. In all its formal as- 
pects The Postman was a scenario the moment it was published. 

Cain has commented on the "legendary success" his stories have 
made as films, saying "I have learned a great deal from pictures, mainly 
technical things." By way of explaining this success, his preface to The 
Butterfly (1947) describes the similarity of all his novels. He writes 
of the wish that comes true, which seems to him a terrifying concept. 
"I think my stories have some quality of the opening of a forbidden 

box " The reader is carried along, he says, by the realization that the 

characters cannot have their wish and survive. 'Thus, if I do any glanc- 
ing, it is toward Pandora, the first woman, a conceit that pleases me, 
somehow, and often helps my thinking." 

All of Cain's novels stylize the Pandora legend. The trespasser ea- 
gerly tells how he did it, then ends up lamely saying, "I see it all clearly 
now, but I don't know why it happened." The hero says this just as he 
is about to be punished. This pay-off makes everything square in the 
movies, but it cannot camouflage the fact that the character is only a 
mechanical monster. 

An author's discussion of his own work is always welcome. But it is 
not the most reliable document in the world, especially when the 
work itself contradicts his own appraisal of it. For example, Cain's 
preface to The Butterfly states: "I care almost nothing for what my 
characters look like, being almost exclusively concerned with their in- 
sides." By "insides" Cain obviously means those inner motives and 
responses which make a character vital. But his characters are almost 
never vital. They almost never have any "insides" except the literal 
kind which occasionally and violently spill out. 

All but one of Cain's novels are the same: a narrative without an at- 
titude becomes a confession without a theme. All that change from 
one story to another are the superficial authenticity of a setting and 
the bewildering virtuosity of each hero's special knowledge. Be it op- 
eratic singing or distilling corn liquor or dieting or picking lemons or 
drilling for oil or mining silver or packaging frozen foods, each skill is 
relevant to the crime. In a way, these books are accurate. 

Past All Dishonor (1946) has the accuracy of an absolute vacuum. 
It tells how a man's passion for a whore leads him to blackmail, steal 

1 28 The Denatured Novel 

and murder in order to get and keep her. The first thirteen chapters 
tell how he goes about it and the fourteenth tells how he gets paid off. 

The chapter sequence would indicate that Cain has already given 
this novel the "treatment." ( i ) A man and woman meet and love; (2 ) 
she leaves; (3) he finds her and discovers she is a whore; (4) he sees 
her professionally and she challenges him to pay the price; ( 5) he earns 
money to meet the price, but loses it before payment; (6) succeeds in 
earning more money; (7) but gets shot in the process; (8) so she nurses 
him but will not accept his marriage proposal; (9) he hires himself 
out as a legal gunman; and (10) kills her wealthy fiance, which (11) 
makes her realize how much he loves her. (12) She responds to him at 
the bottom of a deserted mine shaft, after which he discovers a lode of 
silver ore; and together they plan to steal the money to buy the mine. 
(13) Despite elaborate preparation their train robbery miscarries, and 
they become fugitives; (14) in their hideaway, ever alert for pursuers, 
he shoots and kills her by mistake and so ends his own life there. 

Cain wrote on the dust jacket: "I have tried to put real human be- 
ings before the reader, to explain, as plausibly as I can, how a gunman 
got that way, what the prostitute was doing there, why the mine-owner 
was a bit of a heel, and so on." It is painful to think that Cain could 
have so condescended as to assume his readers would believe this. He 
must have been talking about some other book. 

A novel accretes meaning by its accumulation of analogies, but the 
only thing Past All Dishonor accumulates is irrelevance. The hero kills 
the heroine's finance because this unfortunate man has bought her 
"like a prostitute" as the hero admits she is. He joins the Union Army 
to escape punishment, then goes over the hill with the excuse that he is 
really a Confederate sympathizer. And having already decided to rob 
the train so that he and the heroine can live it up, he announces that 
his theft will keep the treasure away from the Northern Army. 

These tailor-made rationalizations would fit only a robot, and the 
last chapter proves that they do. The hero says he has been sitting 
around for two days and a night: "Writing down how it came about 
that a boy that went to St. Anne's in Annapolis, and believed what he 
heard there, should turn into a traitor, a killer and a thief. I don't know 

Hollywood Pay-Oft 129 

why." So he begins to speculate. "Falling in love with Morina, that had 
something to do with it. But Virginia City had something to do with 
it too." Then he really gets deep. 

Maybe they were wrong about the devil, maybe he didn't move out 
like they said he did. Maybe they just thought he did. Maybe he found 
a new way to conjure. Maybe he found if you give people everything 
they want, and nothing they ought to have, that'll wind them up in 
hell, too. Anyhow, for me it's all over. 

Cain said the movies never bought this book because he had made 
the heroine a prostitute. This seems odd, since movieland can tempo- 
rarily accommodate just about any heterosexual trespass so long as the 
trespasser will be killed anyway. The author explained himself in order 
to emphasize that he had never toned down a novel in order to court 
the favor of a motion picture producer. He must be right in his reason- 
ing about the prostitute, since this novel is constructed exactly like 
every book he did sell to the movies. 

They all rely on the same "switcheroo." In "Double Indemnity" (in 
Three of a Kind, 1943) an insurance claims-adjuster commits a perfect 
murder, and might have collected the insurance money of the deceased 
but for his falling in love with the victim's daughter. This dilemma he 
solves by suicide. Also in Galatea (1953) the narrator tells how his in- 
volvement with another man's wife precipitates violent death. This 
time the husband is the criminal, and the narrator tells of virtue re- 
warded as well as villainy defeated. Discovered and pursued by the 
husband, who is killed in his attempt to kill them, the lovers escape. 
But the lovers must stand trial for his murder, 

The sudden reversal of Galatea takes place in court. Just the oppo- 
site from the trial scene in The Postman (where perverted evidence 
secures the wrong verdict), the lovers are acquitted because the incon- 
sistencies of the truth manage to contradict the circumstantial evi- 
dence against them. This use of the "gimmick" recalls an earlier con- 
fessional, "The Embezzlers," published in Three of a Kind, also about 

1 30 The Denatured Novel 

the narrator's involvement with the innocent wife of a criminal hus- 
band. In this one, however, the hero's innocence becomes so hopelessly 
compromised that he must be saved by an act of author. 

No matter what the twist, it always pays off in the same way. The 
hero of Serenade (1937) has a liaison with a Mexican woman until she 
kills the man who has loved him. Since the pay-off eliminates the mur- 
derer who has eliminated the pervert, the novel excuses everyone's in- 
dulgence in the lurid details along the way. This is what satisfies movie 
values. The message on the paperback cover of the novel announces 
that "Cain is a story-teller first and last everything else in his work is 
stripped to the basic, cinematic action." 

No nonsense here about the novel's complexity. The inside blurb 
of i95opaperbond edition announces that Serenade "has the inevitable 
toughness, the swiftness of pace, the action stripped to essentials. . . ." 
But the novel belies the claim. Precisely because "the action is stripped 
to essentials" there is nothing "inevitable" about it. What is "stripped" 
is the subjective existence of the characters, the thought and feeling 
which fashions all response and gives it significance. 

It is Cain's stylization, not the Pandora legend, that fails; there is 
nothing about Pandora that forbids thought and feeling. In fact, both 
the best and the least of Cain's novels stylize the same legend. One is 
believable and the other is not. The least of these novels, Loves Lovely 
Counterfeit (1942), tells how a stool-pigeon uses the influence of his 
mistress to double-cross the racketeer who employs him, so that he 
can become a racketeer himself. Like the narrator of Double Indemnity 
this hero gets trapped by falling in love with the right woman at the 
wrong time. The irony fails because the hero is not believable anyway. 

Cain later explained (in The Butterfly preface) that Love's Lovely 
Counterfeit was the only novel he had written with "any expectation of 
pleasing" a motion picture producer. "I thought, and still think, Love's 
Lovely Counterfeit ... is a slick plot for a movie, and I executed it well 
enough." It is slick, all right. The novel's final "switcheroo" involves 
the dying criminal's attempt to make his confession outside the juris- 
diction of the law to beat the rap but to satisfy the movies. But even 
the movies would not buy this one. 

What makes The Butterfly (1947) the best of Cain's novels is that 

Hollywood Pay-Off 131 

the hero has believably fooled himself about his trespasses. Brought up 
in the hill country of Kentucky and West Virginia, he preaches the 
literal justice of the Old Testament. The deeds he performs in token of 
his fundamentalist faith, however, are sins by the same token. This 
novel is more persistently sordid than any of the others, involving mul- 
tiple possibilities of incest, bastardy, cuckolding and murder. But 
when this hero dies, with the usual confession, he is believable because 
he has agonized all along over his impossible compromise with God. 

The only one of Cain's novels which explicitly preaches the justifica- 
tion of its hero is also the most immoral. This is The Moth (1948), 
which tells of the hero's twenty-three-year exile from home. During 
these wanderings he has committed armed robbery and escaped un- 
punished; but this is only the half of it. He had originally left home 
under the accusation of seducing his fiancee's little sister. As it turns 
out, however, his innocence of the seduction somehow countenances 
the unpunished crime which he later did commit. 

The last four chapters of this novel which go through the motions of 
trial and sentence would pass for a parody of expedience. They are also 
a good blueprint of Cain's fiction. The prodigal son and his father 
agree on a plan whereby the son will write down his experiences, which 
the father will then edit and evaluate. "I've no gift for words, Dad," 
says the hero, "I'd tell the what and leave out the why." which the 
book has already done for twenty-six chapters. 

Then the father pronounces a long judgment, the heart of which is: 
"What, after all, has loused you, as you put it? Me? Yourself? Circum- 
stances? All three to some extent. But mostly the time in which you 
Jived a calendar." whatever that means. But the author gets away 
with this double-talk because it excuses one crime on the strength of 
the hero's innocence of another. As in so many of Cain's novels the 
moralizing is an accessory after the fact the real fact having been an 
unbelievable character in the first place. 

It seems incredible that Cain could write: "I care almost nothing 
for what my characters look like, being almost exclusively concerned 
with their insides." And from his preface to The Butterfly it is apparent 
that he thought he had not sold out to the movies. His novels, how- 
ever, do not support his opinions. 

1 32 The Denatured Novel 

Perhaps Cain is like most writers who have learned how to satisfy 
the movies and have convinced themselves that the film is really more 
persuasive than the novel. This has fooled a good many novelists be- 
fore him. Back in the early days when Hollywood was beginning to buy 
novels, Jack London mouthed this prejudice. "Visualization is every- 
thing for the teacher/ 7 he was quoted as saying in Moving Picture 
World (January 31, 1914). "I love to teach, to transmit to others the 
ideas and impressions in my own consciousness." 

So far so good. London had just seen the movie of his novel The Sea- 
Wolf (1904). But even as London explained himself, he lost out. "I 
am a realist and essentially a picture writer. In writing a story I always 
keep in mind these two motives; first, I want to make the details so 
plain that he who runs may read, and then there is the deeper underly- 
ing psychological motive." 

Notice the afterthought. This is just what James M. Cain's last 
chapters unwittingly burlesque. As for the priority of motives, "that 
he who runs may read" determines the filmed adaptation of most 

Ever since Jack London's opinion, some forty years ago, the scenario 
method of fiction has continually proved the congeniality of novel and 
film at the novel's cost. The movie reduces the complexities of a 
novel to a conflict, then counterfeits the complexities by visual tricks. 
When these tricks are used to justify the expedient morals of popular 
amusement, then the novel becomes something else again; it becomes 
just another story. 

Chapter VI 

Instant Fiction 

INEXPENSIVE paperbound books, or "paperbacks," now comprise the 
largest share of the book trade's retail volume. Even in a business notor- 
iously built on exceptions and contradictions, the paperback stands 
out. The presentation of the inexpensive paperback customarily con- 
tradicts the book itself; it negates the distribution system which de- 
livers the book to the buyer; and it impugns the very claims it makes to 
the purchaser. Even the marketing of the paperback novel denies the 
reading process; in fact, the publishers have made it possible to not 
read almost every paperback novel they print. 

They have achieved this extraordinary feat by means of the package 
itself, which has created a startling image of American fiction during 
the past twenty years. This image is easy to see. I find it, for instance, 
in the packaging of several hundred novels 302 to be exact pub- 
lished in soft covers during the last twenty years. This is about seven 
per cent of the more than 4,000 fiction titles now in paperback editions. 

There is nothing scientistic about this shelf-list of books (or even 
scientific); it is no controlled sample from specified retail outlets. It 
contains merely the novels I have bought from time to time with small 
change. But it does happen to represent ninety-seven American writers, 
whose works were originally published during the last hundred years; 
and it includes the imprints of twelve major American publishers of in- 
expensive paperback books: Ace, Avon, Ballantine, Bantam, Crest, 
Dell, Gold Medal, Permabooks, Pocket Books, Popular Library, Pyra- 
mid and Signet. It is a fair representation of contemporary American 

1 34 The Denatured Novel 

The image created by these paperbacks is primarily on their covers, 
conveyed by a kind of "outside idiom" which comes across in a hurry 
for efficient non-reading. Just seventy-one adjectives describe all of 
these books, but by a declining ratio. There are forty-nine adjectives 
on the first hundred covers (arranged alphabetically by author), seven- 
teen new ones on the next hundred, and only five new adjectives on 
the covers of the rest. 

A frequency list of the words is even more revealing. Twenty-three 
adjectives are repeated more than five times, and a dozen words are so 
insistent that the image of this fiction is incontestable. These novels 
are "moving," "brilliant," "dynamic" and "compelling." They are 
"gripping," "graphic" and "fascinating." They are "emotion-charged" 
and "blood-stirring," and occasionally "drum-tight." But more than 
anything else they are "poignant" and "powerful." They are about 
equally "compassionate" and "brutal"; and some often the less ex- 
pensive ones are both. 

These adjectives keep company with picturesque verbs. These novels 
were not written, for instance by and large they were "carved." They 
were "carved from the frontier," they were carved "from the city" and 
"from the past," they were "carved from the human heart." These 
carvings are about persons "caught in a web" usually a "web of des- 
tiny," occasionally a web of "stark destiny." Moreover, these novels 
"score." They score by being "merciless," "stinging," "biting" or 
"fearless." And they "speak out" too. But for every "outspoken" novel, 
seventeen are "frank" and three are "utterly frank." The subject of this 
candor is variously "bawdy," "daring," "intimate," "racy," "stark" and 
"unnatural." Searching for a new sensation, one publisher has even 
labeled the contents of his book as "TURGID!" which is certainly 
being "utterly frank." 

What one discovers about these "emotion-packed," "action-charged" 
books by Mark Twain, Herman Melville, Mickey Spillane, Frank 
Norris, Theodore Dreiser, Erie Stanley Gardner, Thomas Wolfe, John 
Marquand, William Faulkner and ninety-one others is that they are 
apparently all the same: a fact, were it true, which would save everyone 
a great deal of time. How all these novels suddenly got to be the same 
in the last twenty years is something of a mystery, until you begin to 

Instant Fiction 135 

inspect the economics of paperback publishing. Then it all becomes 
clear along with the fact that the reader's good faith and the book 
trade's ultimate good intentions have been contradicted in the paper- 
backs more efficiently than ever before. 

These inexpensive editions are the latest version of an old publish- 
ing idea: cheap books in large volume. This newest version dates from 
1939, when Robert DeGraff began the Pocket Books series with thirty- 
four titles whose sales totaled five hundred thousand copies in the first 
year. After World War II the market for cheap books increased in 
every way: in new titles, in gross sales, and in the number of authors 
and publishers represented. After the notoriety of a Congressional in- 
vestigation of pornography in 1952, the steadily increasing sales of the 
paperbacks made good news copy. 

In 1953, when some twenty publishers sold more than two hundred 
and fifty million copies of more than one thousand titles, the nation's 
press began to acknowledge what everyone called the revolution in 
books. Whether one pointed with pride or viewed with alarm, it be- 
came fashionable to talk about books as a mass medium, and to ponder 
the "cultural gain" or loss which resulted from it. In 1955 current 
titles were first cataloged. There are now more than six thousand 
paperback titles in print. As early as 1956 the best estimates of the 
size of this revolution placed the cumulative sales at two billion copies 
of more than twelve thousand titles. The bookkeeping has not yet 
caught up with the facts. 

Only a few facts are needed, however, to assess the effect of the 
paperbacks on the novel and its readers. At first these editions were 
reprints, particularly of fiction. For the first fifteen years of this revolu- 
tion nine out of every ten paperbacks were novels and more than half 
of them still are. During this same period the number of novels issued 
in trade editions, compared to the total number of trade books, has 
proportionately decreased. To all intents, therefore, the paperbacks fur- 
nish the fiction market. 

But as long ago as 1950 the shrinking supply of novels in the public 
domain or on publishers' backlists forced the reprinters to search for 
other sources of supply. They began to publish non-fiction; they pub- 
lished volumes of short stories (a product with which trade publishers 

1 36 The Denatured Novel 

had never succeeded) ; and by a variety of schemes involving both trade 
publishers and film companies, the reprinters began to publish original 
manuscripts. By some lapse of reasoning, however, the publishers dis- 
guised the appearance of these various products in such a way that each 
new book began to look like every other. 

Contemporary publishers have not wholly intended this, any more 
than their predecessors, seventy-five or a hundred years ago, intended 
to ruin the markets which even then existed for cheap books. In the 
1840*5 inexpensive reprints disguised as subscription items became so 
flagrant that Congress adjusted the postal laws and put an end to 
profiteering through low mailing rates. Again in the iSyo's cheap books 
abounded, but with benefit of piracy. For nearly twenty years, until 
Congress passed the International Copyright Act of 1891, the paper- 
bound reprints increased to a point where they totaled one-third of all 
the books produced in the United States. But cheap books miscarried 
again. Even before the copyright legislation, competition had forced 
reprinters to issue such shoddy products that they lost their market. 

This current revival of cheap books exploits some old economic 
conditions in a new way. The paperbacks are distributed not as books 
but as magazines. They are produced consequently like magazines and 
retailed like magazines. They even look like magazines, and the new 
readers they have found are generally the people who used to read 
magazines instead of books. The decline of the pulp magazines after 
the onset of the paperbacks is one symptom of this new readership. 
Another symptom is the high sale, in paperback editions, of novels 
which had never sold very well in the bookstores. 

The most convincing claim of discovering a new audience can be 
made by Erskine Caldwell's paperback publisher. Of the forty million 
copies of Caldwell's books sold to date, thirty-five million have been 
paperbacks; and most of the rest including hard-cover reprints and 
foreign editions were sold subsequent to the paperbound edition 
( 1946) of God's Little Acre (1933; 1946) . This appeared thirteen years 
after its original trade edition (1933), and the reprint sales now total 
more than seven million copies. By means of paperback editions Cald- 
well has sold more copies of more novels than any other writer in the 

Instant Fiction 137 

This makes books a big business or Caldwell's books, at any rate. 
But how many millions make a mass audience? Judging by magazine 
sales, for instance, the paperback reprints have not yet arrived. More 
than six million people buy one issue of Life or Reader's Digest nearly 
as many as have bought God's Little Acre in the past fourteen years. In 
television, sponsors have canceled programs which have reached as 
many as twenty-five million viewers at once. Gross income is another 
way of estimating the mass market, and by this standard the average 
paperback novel is a tiny venture compared to the average Hollywood 

For every book a trade publisher sells, nevertheless, the reprinter sells 
twelve or fifteen. By book trade standards whether by cost, sales reve- 
nue, or number of purchasers the paperbound books are on their way 
to a mass market. Making books for millions instead of thousands has 
forced publishers to face the vexing problem of preserving the book's 
individuality. It is the problem of resisting the compromise of mere 
public acceptance, the compromise of conformity and therefore medi- 

The economic organization of paperback publishing creates great 
pressures toward the conformity of individual books. Reprinting is 
basically a distributing operation, and the distribution of paperbacks 
depends crucially on middlemen who can afford no interest in litera- 
ture except as a commodity. The system really limits the reprint pub- 

The reprinter stands between the trade publisher and the national 
distributor who sells to wholesalers. His contract with the trade pub- 
lisher gives him the right to reprint a title and to sell it in certain mar- 
kets for a limited time at a specified price. For this privilege he pays 
royalties. On a book retailing for fifty cents the reprinter usually 
pays two cents each for the first one hundred and fifty thousand 
copies and a cent-and-a-half thereafter. To secure this contract in the 
the first place he has had to guarantee a larger advance against royalties 
than any of his competitors has offered. 

The reprinter begins, therefore, by assuming some of the trade pub- 
lisher's risk. He prints his own edition, realizing that no matter what 
his break-even figure, he must sell at least thirty thousand copies a year 

1 38 The Denatured Novel 

in order to make up the expense of distributing the book. He then con- 
signs the book to wholesalers who warehouse the book and distribute it 
to retail stores. From the sale of a hypothetical twenty-five-cent book 
the reprint publisher will ultimately receive fifteen or sixteen cents. 
Since his cost (for manufacturing, overhead, royalties and shipping) 
runs to twelve cents a copy, he can count on a net profit of three or 
four cents but only after absorbing the cost of unsold copies; and this 
cost is exorbitant. It takes the profit from eight sales to recover the cost 
of one return. The cost of over-sales has put dozens of houses out of 

Everyone who handles the book, from publisher to retailer, works on 
a narrow margin and depends for his profit on a high volume of sales; 
but the risk belongs only to the reprint publisher. This curious state of 
affairs stems from the marriage of convenience which reprint publishers 
and local wholesalers made during the early years of the present revival 
of the paperbacks. The reprinter could not afford to distribute his 
books by mail; and magazine wholesalers, during the War, were short 
of magazines and needed something to sell. Each needed the other, 
but the method of wholesaling magazines had been established long 
before the reprinter came along. The wholesalers did not buy the books, 
any more than he would buy magazines. He would merely perform for 
a fee the service of distributing them. 

The system is even more limiting to the reprinter. Magazines go out 
of date; new issues replace old ones. To the wholesalers nothing is quite 
so dead as the magazine he distributed last week or the book, for that 
matter. If last week's book is still in the retailer's rack when he makes 
his next call, the wholesaler withdraws it in favor of another title, pos- 
sibly issued by some other reprinter. The publisher must therefore 
continually issue new titles to replace slow sellers. This system puts a 
high premium on predictable books. Each new book must offer or 
appear to offer what the last book offered, only more of it. This fact 
conditions the editorial policy of all paperback publishers. 

Each book must appear to have the salable qualities of its predeces- 
sors. Since the book must compete for attention with magazine covers 
as well as with other books, it must also advertise itself on the retailer's 
racks. The advertising is therefore calculated to make the book stand 

Instant Fiction 1 39 

out but conventionally. It must demonstrate the same appeal but 
more blatantly, which explains those few frenetic adjectives on all the 
eyecatching covers, haranguing the customer like a sideshow barker. 
The aim is to shock, but shock is only a symptom of the affliction. 

The real disorder in the paperback venture is the steady demise of 
the publisher's editorial prerogative. (The quality paperbacks, on the 
other hand, retailed as books and not as magazines for thousands in- 
stead of millions of readers are a different matter. ) What has hap- 
pened to the inexpensive paperbacks whether they cost twenty-five 
cents or seventy-five cents is that the means of distributing a new 
book now determines what the book shall be. It is of no use to exhort 
the reprinter to study the needs and tastes of his readers, if, in order to 
stay solvent, he must continue to produce books which imitate other 
books with a proven sales value. 

You can sense how much the editorial responsibility has been viti- 
ated, by recalling the trade publisher's part in this venture. As originally 
conceived, the inexpensive reprint business repackaged and redistrib- 
uted books already published. The trade publisher had already taken 
editorial responsibility for the book, and his part of the deal consisted 
merely of assigning certain rights to the reprinter offering the highest 
down payment. Reprinters still consider it worth while to maintain the 
appearance of this arrangement, but the nature of the transaction has 
changed to the point where the trade publisher (desperately needing 
subsidiary income) now offers a service to the reprinter. The trade pub- 
lisher knows what kind of books the wholesaler can and cannot sell; if 
his present list does not include such fare his next list probably will. 

Ways and means vary. Before accepting a first novel a publisher cus- 
tomarily solicits bids from reprinters, attempting to find a partner in 
the venture. With an author's second novel the publisher will cer- 
tainly consult the house, which reprinted the first one. Literary agents 
play it both ways: offering a manuscript to the trade publisher, having 
first arranged for the reprint rights; or selling it to the reprinter, who 
then finds a trade house who will bring it out or, in trade talk, 
"famous it up." For the sake of appearances you can argue either way 
about this and be right. On the one hand, the trade publisher occasion- 
ally gets desirable manuscripts he would not otherwise have had. On 

140 The Denatured Novel 

the other hand, short of a veto, he wields no editorial influence. In fact, 
the trade publisher becomes the reprinter. 

Appearances aside, it makes sense to share the printing cost of a 
book's several editions, particularly when the cost can be measurably 
reduced by first printing the pages of the paperback, photo-enlarging 
them for the pages of the hard-cover edition, and then printing the 
trade edition by offset. This saves the cost of a second set of printing 
plates. Although stored and later distributed as a reprint, under this 
common arrangement the paperback is literally the original. 

This confusion of traditional distinctions, and therefore of editorial 
responsibility, has been further blurred by the economic organization 
of reprinting firms. In some cases the paperback reprinter is actually a 
magazine company, and this company's distributing organization 
handles what its subsidiary "publishes." Fawcett Publications, a maga- 
zine house, owns Gold Medal books. Reprints bearing the "Popular 
Library" imprint are published by the subsidiary of Pines Publications, 
a producer of magazines and comic books. 

Sometimes the imprint of a book bears the title not of the publisher 
but of the distributing organization, as with Dell books; but the edito- 
rial responsibility in this case actually belongs to the manufacturer, 
Western Printing and Lithography Company, which primarily prints 
magazines and comic books. The control of another firm, Bantam 
Books, is even more complicated. This house was organized in 1945 by 
Grosset and Dunlap, a hard-cover reprinter, and by the Curtis Publish- 
ing Company, the magazine publisher which distributes the Bantam 
products. But one of the partners, Grosset and Dunlap, was already 
jointly owned by four trade publishers and the Book-of-the-Month 

No matter what his auspices, the reprinter has had to deal more and 
more in original manuscripts. The wholesaler's replacement of old titles 
every few weeks creates a continual need for new ones. The supply of 
published novels can no longer meet these needs. Backlists have been 
depleted and the competition for current wares has driven the prices 
up; so, to cut costs and to gain some measure of control, the reprinter 
publishes original manuscripts. "Originals," in fact, now represent a 
third of the titles published by reprint houses, and this percentage will 

Instant Fiction 141 

This situation offers certain advantages to the author. He gets a bet- 
ter contract from the reprint publisher, retaining all rights but those to 
the paperbound edition. He keeps all the royalties too, for there is no 
trade publisher to share them. It also aids the unknown writer, whose 
first novel might not otherwise have been published. A writer without 
a reputation is often a liability to the trade publisher but no problem to 
the reprinter, whose customers are not impressed by reputations any- 

The reprinter has more control over his product when his product is 
an "original." He customarily offers an advance on a fragmentary 
manuscript several chapters and an outline reserving final payment 
until its completion. This is one way of prescribing fiction, although 
publishers never relish discussing this delicate subject. They prefer to 
point out merely that writing novels to order rarely proves to be com- 
mercially successful. "We don't actually commission fiction," says one 
reprint publisher, "because writers are better than we are at spotting 

Say it however you wish, the author of the paperback original must 
cut his cloth to fit the pattern. The theory of bulk sales with narrow 
margins of profit dictates the need for novels which will sell quickly. 
There are specifications for this, and they are fundamentally like the 
requisites of magazine fiction. This is no mere coincidence. By design 
both products sell quickly, by the same means, and to large numbers 
of customers. 

Like magazine fiction, the paperback novel should offer strong story 
quality, with a prominent conflict and a plot which resolves this con- 
flict by some surprising twist. Also in the manner of the magazines, the 
paperback must avoid offending recognized political, religious or civic 
groups; it must avoid esoteric or learned references, and what publish- 
ers condescendingly call "literary" style. Without the sponsorship of 
advertising the paperback can indulge the reader more explicitly in 
violence and sexual activity, although this is often only superficial to 
the novel. The appeal of the paperback, as one reprinter puts it, is 
"basic approach to basic problems." 

Regardless of the book, this is what the reprinter promises the cus- 
tomer. All over its covers, and inside as well, he strains to show the 
book's extravagance. What he accomplishes by this is the limitation of 

142 The Denatured Novel 

each title to the mean level of all the others. The lie is terribly dam- 
aging to the book and to the reader. The random sample of 302 Ameri- 
can novels, which launched this discussion, raises the question of a 
monolithic literature. When a novel succeeds on the retailer's racks its 
cover copy accurately prescribes the specifications for its successors. 

This inflated "outside idiom" actually limits what it describes. The 
language celebrates excess for its own sake. What other message is 
there except the glut of excess on the paper cover of Meyer Levin's 
novel, Compulsion (1956; 1958)? This copy promises "increasing hor- 
ror and suspense" about "the crime of the century"; it proclaims the 
novel and its subject "sensational," "best-selling," "spellbinding," "en- 
tertaining," "shocking" (twice), "enlightening," "fascinating," "graph- 
ic," "absorbing," "gripping," "provocative" and "startling." "Stun- 
ning" which does not appear describes the total effect. 

There is yet more to the matter. The publishers themselves have 
raised the question of literary excellence by plastering testimonies to 
that effect all over their books. Carefully quoted, these claims are at- 
tributed to "critics," which is what a publisher calls book reviewers 
whose testimony happens to suit him. This testimony appears in the 
language of qualified superlatives. "Masterpiece" and "Classic" per- 
form heavy duty, carefully hedged by disclaimers such as "one of the 
greatest" or "among the greatest," "perhaps the greatest of all time," 
or "in the nation" or "in the century." So-and-so, an unknown author 
who cannot yet qualify for greatness of this magnitude appears instead 
as a "master story teller" or merely "at his best." 

What palls is the utter misrepresentation of a book's contents. The 
paperback edition of Ernest Hemingway's novel Across the River and 
into the Trees (1950; 1950) is really a collector's item. It carries Ten- 
nessee William's affidavit that this is "the best ... the finest thing that 
Hemingway has done." But this laughably bad judgment is a mere 
amateur's try compared to John O'Hara's absurd statement, also 
quoted on the cover, that Hemingway is "the outstanding author out 
of the millions who have lived since 1616." When it comes to handing 
out literary credentials no publisher has yet matched this package, but 
they are all trying. 

The affidavit of literary association on the cover is used to excuse or 

Instant Fiction 143 

to justify the book. It also braces up the reader; it is that little extra 
something which helps the tourist pretend he is traveling first class. 
The reprint package of Frederic Wakeman's Shore Leave ( 1944; 1948) 
typically flouts such big-time literary talk. This book has "some of the 
wise-cracking wit of Hemingway's The Sun Also Rises" "It's like an 
almond-bitter Farewell to Arms; a more desperate Hemingway in a 
more tragic age/' About the hero: "All by himself he made up a lost 

The blurb in the paperback edition of Erskine Caldwell's God's 
Little Acre likens the hero's lusting after his daughter-in-law to the 
"Song of Solomon." We are told of Tobacco Road ( 1932; 1947*!) that 
Jeeter Lester is "heroic in his ineradicable love of the soil, his refusal 
to leave it for the superior living conditions of the town and its indus- 
trial civilization." Something has suddenly happened to the concept of 
the hero after several thousand years. Every notion of heroism which 
the modern world has inherited involves awesome power, size or glory; 
and Jeeter Lester violates every one of these qualities. The blurb ac- 
companying Georgia Boy (1943; 1950) similarly patronizes its prede- 
cessors. This is asserted to be a classic of American boyhood which will 
stand beside Hemingway's "My Old Man" and Mark Twain's Huckle- 
berry Finn. But the fact that Georgia Boy has nothing to do with the 
juvenile character who tells the stories dismisses the book from this 
particular hall of fame. 

The book publisher has an impeccable answer to this sort of caviling. 
After all, he did not write these critical claims, he merely displays them. 
They are all carefully quoted. Nor is there anything new about the ex- 
travagance of book reviewing. He is right, of course. 

The low state of book reviewing in America has always been attacked 
by writers, even by reviewers themselves. Nearly 150 years ago John 
Neal interrupted his novels to sermonize on this subject. E. A. Poe 
made capital and good copy of incompetence and knavery in book 
reviewing. The whole matter of literary exploitation so bothered Henry 
Thoreau that in editing the recollections of his Concord River journey 
he inserted the note that "books are for the most part willfully and 
hastily written, as parts of a system, to supply a want real or imagined." 
This sounds like the disdainful comments of Edmund Wilson, Irving 

144 The Denatured Novel 

Babbitt, and O. H. Cheney. Testimonial advertising always attempts 
to manufacture masterpieces where none exist. 

The contemporary reprinter, however, has contributed something 
new to this malpractice an entire tabloid package which distorts more 
efficiently than ever before. Comparison by superlative soon exhausts 
its subject. When such comparison involves two billion packages, more 
or less, then who can tell from these packages what the contents of any 
one of them might be? How shall the reader know his purchase? All of 
these novels look the same a new kind of instant fiction packaged in 
two billion compelling sequels to one another. 

The reprint publisher concentrates on a quick sale to the customer. 
The book must sell itself, must catch the attention and lead it rapidly 
into the story. Its whole presentation builds the habit of buying on 
impulse, and one way to do this is to promise quick sensation. In this 
way the reprinter has achieved a large volume of sales, but not neces- 
sarily large numbers of customers. As every trade publisher knows, sen- 
sation can always be sold to a constant fringe-market, no matter what 
the subject of the book. Despite this, publishers usually find it better 
business to slant books toward the wider market of persons who feel 
more comfortable with the accepted mores. It is just possible, however, 
that the reprint packages which sell sensation only are purchased by 
this same fringe of buyers over and over again. 

Even assuming that they know their customers, the reprinters (like 
the trade publishers) have missed the distinction between book buyers 
and book readers. They have successfully exploited the truism that 
shallow books catch on quickly, or not at all. But by selling all their 
novels as instant fiction, they penalize those books which have some 
complexity. It takes more time to read the novel which demands some 
reasoning collaboration; therefore, it takes such a novel more time to 
be talked about and to sell. But the paperback's entire presentation 
denies this fact. It does so by always encouraging the buyer to expect 
entertainment without effort. 

The presentation usually makes this indiscriminate promise by 
abridging the contents of the novel into a preview, usually printed on 
the book's first page but sometimes on the back cover. This preview 
tells the reader what to look for. It stresses the "becoming" movement 

Instant Fiction 145 

of the novel, stating a conflict and raising the question of how the novel 
will resolve it. The preview insists on some unique excess within this 
conflict which makes it different from everyday experience and gives 
it news value. 

The preview is journalistic. Like the news story, which telegraphs its 
climax, the preview keeps the reader going. It even looks like a news 
story, from the headline and the lead sentence featuring the unique- 
ness, to the narrative of events, and finally to the occasional details. 
This newsworthy preview ideally fits those novels which have nothing 
more to offer than the resolution of a unique series of events. 

The preview in the paperback edition (1953) of John O'Hara's Ap- 
pointment in Samarra (1934; 1 9$3), f r instance, gives a faithful ac- 
count of this novel. After a 24-point headline which proclaims a "Fatal 
Flaw," the lead sentence identifies the hero who passionately loves his 
wife, but whose character is "flawed by a destructive weakness." Three 
sentences then paraphrase the action: 

In the brief three-day span of this tense ironic novel, Julian's life 
moves swiftly to its inevitable tragic climax. His downfall begins at a 
country club dance when he throws a highball in the face of Gibbs- 
ville's wealthiest, most powerful man. A drunken encounter with a 
nightclub singer in a gangster's roadhouse lends its swift momentum, 
and from a disgraceful brawl in a fashionable men's club through a 
final, bitter quarrel with his wife, Julian's life spirals downward with 
dizzying speed. 

Another sentence informs the reader how quickly it all happens. And a 
final statement testifies that the novel is a "modern classic." From 
beginning to end this preview expands its facts like a news story. 

The newsworthy resume stresses the action and disallows its impli- 
cations. But because novels differ from one another precisely by these 
implications, the monolithic preview usually distorts a novel. The 
paperbound edition (1954) of Walter Van Tilburg Clark's novel, The 
Ox-Bow Incident (1940; 1954), for example, distorts by default. In the 
customary sequence of headline, lead sentence, synopsis and "expert" 
literary testimony, it clarifies the story's news value. This is "a powerful 

1 46 The Denatured Novel 

novel of the American West"; it is "an unforgettable tale that blazes 
with the strength and vigor of its setting"; and it is "a searing and ex- 
citing story of honest men of action who let their mistaken fury lead 
them to violence and injustice." 

This generality scarcely suggests the novel's real subject, which is the 
inadequacy of mere virtue in the world. At a kangaroo court, a posse 
of ranchers has tried and executed three innocent men. Following the 
"shocking outdoor trial" where the preview of the narrative stops 
the man who has clearly tried to prevent the execution now astonish- 
ingly takes the responsibility for it. Convinced of the suspect's inno- 
cence, he argues, he has nevertheless joined the posse unarmed, pur- 
posely to avoid a showdown with its leader. With this confession the 
novel then focuses on the other men who have given their frightened 
consent, as they become aware of their own irresponsible righteousness. 
These implications of the execution, which the novel is all about, are 
not even whispered in the preview. 

The paperback presentation of Robert Penn Warren's All the King's 
Men (1946; 1951 ) "A GREAT NOVEL . . . about great corruption" 
similarly overlooks what that novel is all about. The preview features 
the unscrupulous politics of Willie Stark, who forces one of his em- 
ployees to double-cross a personal friend. From the moment he does so, 
"the story moves steadily to its tremendous climax." In this way the 
preview alludes to Stark's assassination. But the novel's real subject is 
the pathology of Willie Stark's power, and it is dramatized through the 
self-realization of a few well-meaning people who, for convincing per- 
sonal reasons, have sold out to this man. 

Self-comprehension is what a novel can best represent, but this is 
just what the previews omit. Any fiction appraises experience simply by 
resolving a conflict, but the tremendous possibility of the novel lies in 
the breadth it has in which to make analogies to this conflict, such as 
the shared problems of the various ranchers and their victims in The 
Ox-Bow Incident, or of the several employees of Willie Stark in All 
the King's Men. The previews commonly emphasize the action for its 
own sake, but this merely advertises a book's similarity to all others. 

If a preview were to emphasize the action with regard to its particu- 
lar implications, on the other hand, it would advertise the book's indi- 

Instant Fiction 147 

viduality. The difference would be more than a matter of altering 
words here and there; it would mean changing the preview's point of 
view to accommodate the book itself instead of the reader's assumed 
tastes. And this is all the difference in the world. 

See how obviously the reprinter limits what the novelist offers in the 
first paperback edition of Nathanael West's Miss Lonelyhearts ( 1933) . 
After a screaming headline ("Dear Miss Lonelyhearts, HELP ME . . ./ 
HELP ME . . ./ HELP ME, PLEASE!!!") the preview informs the 
reader that this is the lovelorn column of a big city newspaper, "and 
when 'Miss Lonelyhearts 7 turns out to be a man the result is bound 
to be both explosive and shocking." But the shock and explosiveness 
have nothing to do with the gender in the title; the novel's opening 
sentence settles that. The point of the book is not that Miss Lonely- 
hearts "turns out to be a man," but that he turns out to be a man with 
a Christ-complex, whose inability to cope with human misery out- 
rages and destroys him. The novel so far exceeds the reprinter's dis- 
torted promise that the preview might better be placed between the 
covers of another book. In any case, it has been dropped from the later 
printings of this one. 

This limiting journalism of the reprinters even extends to tabloid 
reading-guides on the cover itself, to help the reader make up his mind. 
Thus, the retitled paper edition The Night Before Dying (1955) of 
Robert M. Coates's Wisteria Cottage (1948; 1955), saves the reader 

Louisa haughty, love-hungry with claws of steel/ Elinor warm, allur- 
ing, and very, very soft/ Florence shrewd, calculating, and too fond 
of young men/ THESE THREE:/ caught up in the web of rage and 
contempt spun by the twisted mind of Richard Baurie, the youth who 
deceived them all and then, step by terrifying step, directed their lives 
down the path to explosive violence. 

What more do we need? One more step and we can get through a book 
without reading it. The paper edition ( 1952 ) of J. P. Marquand's Point 
of No Return (1949; 1952) admirably performs this final chore in a 
paragraph of illuminated writing, with pictures punctuating the prose. 

148 The Denatured Novel 

Here is Charles Gray, ( ) still dreaming his dreams of his lost, lovely 
Jessica, ( ) living out his life in quiet desperation, with his nice wife, 
( ) and their two kids, { ) ( ) in their Connecticut home. ( ) 
The boundaries of Charley's life are home, the country club ( ) where 
he plays golf, the bank ( ) where he works. 

Another paragraph states but without informing what the hero's 
universal problem really is: 

Here is the poignant story of a man imprisoned in the grip of circum- 
stances beyond his control a man caught in the net of his own daily 
life. POINT OF NO RETURN is a beautiful, powerful and universal 
novel, by one of America's greatest writers ( ) JOHN P. MAR- 

This nonsense is the result of economic cause. The present revival 
of inexpensive paperbacks owes its existence to a particular method of 
distribution, and the books themselves have been tailored to accom- 
modate it. This method of bookselling has caused remarkable changes 
in the form of the novel itself but more obviously in the representation 
of the whole idea of literature. The latter consequence deserves a close 
look, for the reprinter has become a taste-maker. 

Put yourself in his shoes. In order to tell the reader what to look for 
and to condition him for more of the same, you must ignore or mis- 
represent the relative worth of the individual book. This is true even 
if you are publishing the works of a single novelist, particularly if he 
happens to be a best seller. Once committed to an author, once started 
on this market-building program, the publisher cannot countenance 
the difference between artistic success and failure, even though this dif- 
ference may be vast. He must press all these works into the same shape 
and appearance. 

The collected reprints of Erskine Caldwell decisively illustrate this. 
More copies of his books have been sold than of any other novelist. 
Caldwell is one of the monoliths of the reprint business; even by its 
own standards his gross sales are extraordinary. Since 1946 his twenty- 

Instant Fiction 149 

five paperbound novels and short-story volumes have so far sold thirty- 
five million copies. In addition to the seven million copies of Gocf s 
Little Acre four of his other novels have each sold between three mil- 
lion and five million copies; the sales of two others are approaching 
three million; and five more of his titles have sold over a million. 

Caldwell can explain this sales record in his own obtuse way. "My 
grandfather, who had the reputation for telling the biggest whoppers 
ever heard in Sycamore County, Georgia, used to tell me that his stories 
were actually no better than hundreds told by others, but that he had 
learned how to raise interest to the point where people would begin 
nudging one another in anticipation when they saw him coming down 
the street." The advertising copy on his own books has the same effect. 

The blurbs in his reprinted novels and short stories make interesting 
reading. They repeatedly claim for their various contents nothing less 
than the entire aesthetic range at the same time. One novel achieves 
"a matchless gift for portraying comedy and tragedy"; another, "the 
matchless blend of comedy and tragedy"; still another is "funny, 
earthy, tragic." His first book, Tobacco Road, offers "humor, tragedy, 
horror and pathos," and Episode in Palmetto (igSo; 1953 ), published 
twenty-five years later, is a "moving mixture of the comedy and tragedy 
of everyday life." Ths Sure Hand of God (1947; 1949) is a "riotous, 
bawdy, and sometimes tragic story"; and the episodes in Gulf Coast 
Stories (1957) " ran g e from bawdy comedy to quiet tragedy." 

The impression is inescapable that this author offers all things to all 
readers, In fact, this is just what the preview in God's Little Acre as- 
serts. "What accounts for the extraordinary appeal of this book?" the 
copy writer candidly wonders. "The answer, the critics say, lies in its 
humanity which offers every reader the kind of reading experience he 
prefers." Everybody wins a prize. "If it's laughter you like, it's here in 
abundance; if it's social significance, you'll find that too; if you seek 
insight into the lives of people who may seem very different to you, it's 
here. And if you're looking for good writing, Erskine Caldwell is one 
of the best literary craftsmen of our veneration." 

If you sense the splendid equivocation in this marketese you are on 
the right track. In fact, the copy writers, are not alone in their puzzle- 
ment over Caldwell's fiction. Its tone is undeniably ambiguous. The 

1 50 The Denatured Novel 

author disdains public comment about it, by himself or by anyone 
else; and there is no substantial appraisal of it. No one has yet contra- 
dicted these claims. 

There is reason for this reticence. Caldwell's autobiography, Call It 
Experience (1951; 1956), says nothing significant about his writing, 
although his comments in another volume do get to the point. For the 
trade edition of a collection of short stories, called Jackpot (1940), his 
publisher prevailed upon him to write a series of prefaces, and Cald- 
well wrote seventy-five statements ranging from a sentence to a para- 
graph including that one about his grandfather which parody the 
art of prefacing. In its entirety one of these prefaces reads: "If Profes- 
sor Perkins would relax, he might at least get a little fun out of some 
of these stories." And later: "If Professor Horatio Perkins is still leafing 
through these pages, I hope he will not become discouraged and give 
up. Even if he does not succeed in discovering the secret of writing 
fiction, at least he will find comfort in the knowledge that the author 
himself is up a similar tree." 

In one of these prefaces Caldwell explains how he parries polite ques- 
tions about his work. Once a year he jots down ten or fifteen titles, and 
every three or four weeks he writes a story for one of them. It saves the 
trouble of looking around for titles afterward. As this straight-faced lie 
might indicate, Caldwell likes shaggy-dog stories. His fiction is just as 

This sort of remark, like his fiction, puts the critic off. Despite gen- 
erous reviewing, Caldwell's fiction has received only scant critical atten- 
tion, for two good reasons. His fiction has made a great deal of money, 
which tends to embarrass academic criticism; and it is often funny 
terribly funny but without any obvious significance. His stories are 
not satires; they encourage no social program; and such undirected 
humor embarrasses academic spokesmen. Caldwell has even affected 
dismay over the interpretation of his writing as humorous, thereby 
adding another hair to the shaggy dog. 

Erskine Caldwell writes fantasies about improbable people. He 
writes about grotesques who understand one another but who are out 
of touch with the rest of the world, like the crackers and the hillbillies 
he grew up knowing. They are incongruous characters: they take irrele- 

Instant Fiction 151 

vant matters seriously, and they take the momentous for granted. They 
somewhat resemble the characters in medieval fabliaux, such as those 
Chaucer's Reeve and Miller and Summoner tell about, or the rogues 
in Fielding's or Smollett's novels. But Caldwell's fiction has more of an 
animus, like the tall tale which insults the listener, such as Mark 
Twain learned from his newspaper colleagues. 

Caldwell caricatures these peculiar people. Except for a few bold 
details the reader must complete them for himself and make his own 
assessment. The author merely tells anecdotes about them. He dis- 
closes, without developing, a situation. He writes in Journeyman (1935; 
1947), of the man who bets his father-in-law's watch in a crap game, 
covers it with his automobile, then his wife and loses them all. He 
writes of a father, in Tragic Ground ( 1944; 1948 ) , who calls at a brothel 
to take his daughter home, forgets his errand, and stays the night. He 
tells of the fun-loving widow, in The Sure Hand of God (1947; 1949 ), 
who entertains her gentlemen callers with a hypo of what she believes 
to be vitamins. The implications of these anecdotes, however, become 
the reader's responsibility. 

These are typical anecdotes, bizarre and toneless. The quality of 
every one of his novels depends on the extent to which such anecdotes 
become analogies, or to which they develop the implications of one 
another. In one of his prefaces to Jackpot Caldwell writes, "I have a 
great respect for my grandfather's method of criticism. To him a story 
was either a humdinger or God-awful." And these are precisely the 
alternatives of Caldwell's own novels. 

The least of his novels are merely literal repetitions of anecdotes. In 
the best of them, however, the anecdotes variously enforce one another 
and evolve a meaning beyond any one of them. By literal repetition 
Caldwell's improbable people become impossible, inert characters. 
But when the episodes vary, they suggest a whole world in which these 
improbable people come alive. This distinction marks Caldwell's nov- 
els far more fundamentally than the superficialities of subject matter 
on which the copy writers continually harp. 

Trouble in July (1940; 1945) is Caldwell's finest novel; This Very 
Earth (1948; 1951) is one of his worst. Their subjects are utterly dif- 
ferent. But you would never guess these differences from the paperback 

152 The Denatured Novel 

blurbs. The news-story preview of each novel promises the display of 
some sociological problem: one of race hatred and the other of juvenile 
delinquency. The prose of both blurbs is charged with hormones. One 
teaser features the dishonesty of "an over-sexed, promiscuous girl," 
and the other features a girl's vexing hunger for a man, "to keep from 
being lonely.' 7 The advertising marvelously disguises the facts. 

Trouble in July tells of a man-hunt which ends in two murders. One 
is the lynching of an innocent Negro accused of raping a white girl. 
The other is the execution of this girl who has wrongfully accused the 
Negro. The lynchers stone her to death. The horror of this story is its 
believability. These obsessed characters do come alive, as the novel re- 
veals a state of mind that prevails in their world. Their awesome stu- 
pidity is incongruous with any outside sense of values; but the wild, 
fanciful anecdotes which comprise the novel have so diversified brutish 
irresponsibility that these characters seem capable of anything. 

The anecdotes complement one another. The irrelevant indulgences 
of the characters can be ludicrous and laughable or terrible, depending 
on the circumstances. The fat sheriff "wants to keep the lynching po- 
litically clean" to avoid publicly declaring himself for this Negro 
against the white men whose votes he will need. But he also wants to 
keep up appearances, so he locks himself in jail on the fatal evening to 
make it seem as though the lynchers had forced him out of action. It 
develops, however, that in the dark he has locked himself in the wrong 
cell, with a mulatto woman in whose company his wife and the lynch- 
ing party later discover him. 

Another episode concerns the argument between the village barber 
and Shep Barlowe, the girl's father, over how to conduct the lynching. 
They start to fight with knives, and when an old man tries to restore 
the peace they knock him down and presumably kill him. The old 
man's fate is not clear, however, since no further mention is made of 
him. Still another episode recalls how Shep Barlowe behaved when his 
wife died. She disappeared one day, upsetting his whole routine, and 
his temper lasted a week before he set out to find her. When he finally 
did discover her corpse at the bottom of the well, he threw all his cord- 
wood down after her. 

No editorializing accompanies these anecdotes, but they comple- 
ment one another. They develop the conviction of a kind of mass mind 

Instant Fiction 153 

that can hang or stone or bury in cordwood anyone who opposes it. 
Even so, the Negro might have escaped had he not stopped to feed his 
rabbits; and the girl wishes she could bring herself to tell the men 
about her mistake. Thus, the novel cumulates the sense of monstrous 
incongruity peculiar to a lynching. 

The sadism in This Very Earth is no more costly, no more horrible. 
It is just not believable. The reason is that this novel's anecdotes liter- 
ally and categorically repeat one another, and nothing relieves the con- 
stant extremity. In this book a man named Chism Crockett has re- 
signed his life to the indulgence of his own passion, which includes the 
exploitation and the torture of his children. He gets his eleven-year-old 
son drunk and makes the boy watch him fornicate. He blackmails one 
of his daughters and tries to exploit a love affair of the other. The man 
has no more motive than the white girl has in Trouble in July. But this 
novel lacks the various disclosure that might imply a cause. 

Chism Crockett merely and literally repeats himself throughout This 
Very Earth, and so do the other characters. Chism's son-in-law is seen 
almost exclusively in the act of beating his wife (the only development 
being that he finally beats her to death). Chism's son and Chism's 
father are also caricatures; they also exist without living. One repeated- 
ly tries to reason with Chism and the other repeatedly harangues him. 
All the characters affect concern over whether Chism should stay in 
the city or move back to the country so that his children can have a 
wholesome life as though this had anything to do with his behavior. 
These characters exist only by the author's assertion, and each one 
compounds the felony by imitating himself. There is no cause or con- 
sequence in their actions, and no reason in the novel. 

By insisting on the sensations of one of Chism's daughters the 
"Bad Girl" who is "too free and easy ... to keep out of trouble" the 
preview in the paperback edition of This Very Earth suggests a ra- 
tionale the novel does not have. This is true of the presentation of most 
of Caldwell's novels. The advertising insists on the vitality of sex or sin, 
but the novels disclose no such vitality. The fiction contains characters 
who literally repeat their own excesses, and this goes unassessed by any 
other behavior. The fault is not in the sexuality but in the unbelievable 

Once, long ago, Caldwell did build a convincing novel about this 

1 54 The Denatured Novel 

goatish concupiscence, In God'; Little Acre the members of Ty Ty 
Walden's family live and die by lust. The beautiful Griselda inspires 
Ty Ty's embarrassingly funny rhetoric, her brother-in-law's animalism, 
an absurdly futile pursuit by another in-law, and a murder by her 
jealous husband. These irregularities cause Ty Ty continually to la- 
ment God's unfortunate error in creating animal and human in the 
same being, and Ty Ty's laments reveal the shrewd stupidity by which 
these people justify themselves. 

Ty Ty's denial of his tithe, by moving God's Little Acre expediently 
out of the way of each prospective gold mine on his property, drama- 
tizes the fallibility of these characters. He has so often violated the 
Third Commandment that when the murder happens to occur on 
God's Little Acre, lately removed from the site of the newest dig, the 
coincidence becomes a retribution. By their various distortions these 
ludicrous anecdotes develop an idea about the rationalization of hate 
and desire which can put an end to everything. 

But this was many years and many novels ago, and as each new re- 
print reminds the purchaser that it is written "by the author of God's 
Little Acre," the irony in the gross implication becomes more painfully 
obvious. The case of Erskine Caldwell is typical. Every reprinter prom- 
ises more of the same, no matter who the author or what the book 
happens to be. 

There is an old joke in the book trade about the person who asks, 
"Why should I buy a book? I already have one/ 7 But this turns out to 
be a pretty expensive joke. The reprinters' distortion of their products 
has encouraged just this sort of inertia. With the narrow margin of 
profit, every reprinter understands the vast effect of a small change. A 
half cent a copy, even for one printing, can mean the difference of 
thousands of dollars. The reprinter is used to this kind of arithmetic 
and to the need for precision, but he overlooks the comparable effect 
of small changes in the literary quality of his product. This is a curious 
oversight, since he himself makes such pretension of displaying a 
book's literary credentials. 

It is irrelevant to ask why the reprinter does not publish better books. 
He publishes the best he can find, commensurate with the risks of a 
large-volume business. In this particular matter of conscience he can 

Instant Fiction 155 

probably sleep better at night than the trade publisher. But by market- 
ing his wares all as the same product he distorts the books he does pub- 
lish. Every novel is the extension of its author it is individual. The 
value of a novel lies in this significant difference, within the experience 
everyone shares. But the reprinter denies this. 

The parrot-like salesmanship of the reprint business contradicts it- 
self. It denies the possibilities of the distribution system. Since the 
basis of the business is the rapid turnover of many titles with very 
little display room in which to sell the reprinter must attract atten- 
tion to each book. To this purpose the individuality of a book is natu- 
rally congenial. It is even strategic in finding new readers. But by ignor- 
ing or disguising this individuality the reprinter encourages the reader 
to limit every book that he does buy to what the reprinter says it is. 

Meanwhile the image grows of a literature so parochial that all its 
books render the same experience, that all are equally "poignant" and 
"powerful/' "brutal" and "compassionate," and that one can do the 
job as well as another. 

Chapter VII 

After Marquand, the Deluge 

IF A NOVEL convinces readers of what they want to believe anyway, 
chances are it will succeed commercially. If it informs or instructs the 
reader at the same time then it can be readily excused for being fiction 
and for being successful. This takes some doing. 

The image of such fiction of an entire literature of such fiction, in 
fact is what the inexpensive paperbacks attempt to sell. But paper- 
back advertising merely extends to more books the implication of quali- 
ties magazine fiction and movie fiction actually possess. The magazines 
and the movies have stylized the novel form to make it serve acceptable 
attitudes without seeming to. 

The fiction in both these media stresses a sharp, definable conflict 
which will surprisingly resolve in a favorable manner. To help disguise 
this fortuity, to make it seem convincing, this fiction offers a superficial 
authenticity of costumes, setting, speech and manners. More impor- 
tant, such fiction contains no analogies which would test or evaluate 
the conflict. A single encounter exists by itself, solved or "proven" in its 
own terms, with all the absolute value of a cat's-cradle. 

These are permanent qualities of the stylized fiction of the maga- 
zines and the movies. Making the illusion convincing is a matter of 
making the conflict localize some large, public, and usually urgent 
difference of opinion; and there are as many ways of doing this as there 
are writers. But there is usually a similarity in ways and means from 
one fiction to another. 

This is a sore point with all writers. Although there is always a con- 


After Marquand, the Deluge 1 57 

tinuity of ideas and techniques in any art form (no writer can achieve 
what he does without others who have gone before him), any author 
resents it when his work is made into some facile equation with the 
work of others. In his preface to The Butterfly James M. Cain spoke for 
the majority of writers when he said, "Schools don't help the novelist, 
but they do help the critic; using as the mucilage the simplifications 
that the school hypothesis affords him, he can paste labels wherever 

convenience is served by pasting labels " Cain straightened out the 

critics, "these strange surrogates for God," in this elemental matter. 
"You're really being a little naive, you know. We don't do it that way. 
We don't say to ourselves that some lucky fellow did it a certain way, 
so we'll do it that way too, and cut in on the sugar. We have to do it 
our way, each for himself, or there isn't any sugar." 

The novelist writes as he is, no doubt. But that is only half the 
story. His work is published under auspices which deny this spirited 
principle of individuality. "Imitation by publishers has become an in- 
creasingly frequent cause of intensified competition," O. H. Cheney's 
Survey informed the trade in 1931. In a chapter entitled "The Book, 
The Buyer, and The Critic," Cheney wrote of this compulsive publish- 
ing which causes disastrous overstocks: "two or three simultaneous 
books on some English poet or French beauty, seven simultaneous 
biographies of a football coach, hordes of books on Russia. Even if 
everyone of the books were good, what good would it be?" 

This is not just a problem of quantity. By a calculated imitation the 
book trade sells books by disguising whatever individuality they have. 
Publishers think of books in categories and exploit any association they 
can. A book's entire presentation is usually aimed at cashing-in on 
some previous success. This is almost a matter of superstition among 
publishers. It even extends to the titling of books. 

When Alfred Knopf published Kurt W. Marek's exciting study of 
archeology (under the pseudonym of C. W. Ceram) and entitled it 
Gods, Graves, and Scholars (iqSi ), it sold startlingly well. When they 
were sure of its success other publishers moved in. Knopf had appar- 
ently uncovered more of a cache than his archeologist had. The Univer- 
sity of Oklahoma Press happened to release a study by Joseph L. Smith, 
entitled Tombs, Temples 6 Ancient Art (1956). But the three-decker 

1 58 The Denatured Novel 

alliterative title had already been rediscovered in Leo Gurko's Heroes, 
Highbrows and the Popular Mind (Bobbs-Merrill, 1953) and Paul 
Wellman's GZpry, God, and Gold (Doubleday, 1954). 

"God" was a frequent partner in this multitude of success stories. 
Henry Armstrong's autobiography, Gloves, Glory, and God (F. H. 
Revell, 1956) was followed by Howard E. Kershner's God, Go/d, and 
Government (Prentice-Hall, 1957); by George Ashbaugh's story of 
the Mormons, Gods, Sex, and Saints (Augustana, 1957), and by 
W. H. D. Rouse's Gods, Heroes and Men of Ancient Greece (New 
American Library, 1957). Prentice-Hall was still at it with the title of 
William E. Hulme's book, God, Sex & Youth ( 1959) . 

If a three decker title cannot accommodate "God," then the next 
best thing is alliteration. Rogues, Royalty, and Reporters (Hough ton 
Mifflin, 1956), by William B. Ewald, appeared in the same year as 
Zoomies, Subs, and Zeros (Greenberg), by Charles A. Lockwood and 
Hans C. Adamson, and Magic, Myth, and Medicine (World), by 
Donald T. Atkinson. An anthology of crime, Murder, Mayhem and 
Mystery (Barnes, 1958), edited by Alan Hynd, followed them to mar- 
ket, along with Blondes, Brunettes, and Bullets (McKay, 1957), ^y 
Nils T. Granlund. 

Even if publishers could not work in "God" or alliteration, there was 
still the magic trinity of Gods, Graves, and Scholars. Louis Paul's 
Heroes, Kings, and Men (Dial, 1955) an( ^ Grillot De Givry's Witch- 
craft, Magic and Alchemy (Books, 1958), attest to this. And if one 
trinity is not enough there is always the sub-title, as demonstrated by 
William Irvine's Apes, Angels, and Victorians; The Story of Darwin, 
Huxley, and Evolution (McGraw-Hill, 1955). Real originality fright- 
ens most publishers, to the detriment of theirbooks. There is more 
than just the reassuring likeness of these titles. Most of the books 
ended up being hawked for a few cents in the mail catalogues of the 
"remainder" houses. 

This is the way publishers herd their products. When a book does 
find a market it will surely be followed by others like it or as nearly 
like it as their publishers can make them seem. This explains the do- 
mestic novels which clustered after Susan Warner's The Wide, Wide 
World (1850), the novels of ethical religion which flocked after Rob- 

After Marquand, the Deluge 1 59 

ert Elsmere (1888), or the vogue in fat historical novels which An- 
thony Adverse (1933) and Gone -with the Wind (1936) revived. But 
these are merely some obvious cases of what goes on all the time. 

Whether an author belongs to any "school" does not much matter 
his publisher will enroll him in one. A successful book has always 
created a market for more like it, and there are always writers willing 
to gratify the demand. The evidence of these clustered books is always 
the same; their mannerism deteriorates the original appeal. They cari- 
cature those qualities in the original which were presumed to make it 
sell in the first place. Sometimes the mannered imitations even carica- 
ture one another. 

There are different ways to describe this syndrome. Alfred Knopf has 
described it in terms of the difficulty of publishing the works of new 
authors. "One of the reasons ... is that they have to compete dollar 
for dollar in retail price with works of a similar kind and length by the 
best-established authors." (Knopf printed this in The Borzoi Quar- 
terly, Second Quarter, 1957. ) As he put it, it is a case of the reader buy- 
ing a sure thing or buying a risk. "In other words you can buy five or 
six hundred pages of a new novel by John P. Marquand and know just 
exactly what you are going to get, or pay the same price for a book of 
the same length by someone you never read and have never heard of, 
which book may utterly disappoint you after fifty pages." One expedi- 
ent remedy, of course, is to minimize the risk, to make the new novel 
sound like one of John Marquand's. This would be mannerism. 

Wherever yon find fiction directed toward a timely subject you will 
find mannered novels. A good bet for a timely subject is business and 
the businessman; and most business novels display obsequious man- 
nerism. For the past ten or twelve years the magazines have given 
business novels devoted attention. Every few months pedants gather, 
sort, and classify attitudes in novels to discover whether or not the 
author really approves of his hero. The polls have nothing really to say 
about fiction, since fictional attitudes cannot be classified as though 
they were actual. Only rarely can readers agree on thematic niceties in 
any competent and reasonably complex novel. Aside from what they 
attempt to prove, however, the polls do reflect a lively interest in the 
business novel. 

1 60 The Denatured Novel 

It is not a new but a continuing interest. American business novels 
began to appear after the Civil War, when it looked as though indus- 
trial capitalism would destroy the ideology of an inherited democratic 
and agrarian way of life. Machines, technological unemployment, and 
the pressures that heavy capital could exert apparently threatened to 
annihilate free enterprise. Some novels dramatized the moral problems 
of making money; some exposed business malpractices; a few worried 
over the threat of a powerful proletariat; most of them localized the 
love-and-duty conflict in a business setting and steered clear of ideas. 
But right from the start American business novels capitalized on the 
public sense of urgency about their subject. 

By 1900 novelists could count on the widespread distrust of business, 
particularly of Big Business. The many-sided transactions of financial 
capitalism had presented new opportunities for fees, confused the 
notion of profits, and increased the suspicion that the middleman was 
growing disproportionately rich. The Congressional investigation of 
the money trust, in 1912 and 1913, consequent to J. P. Morgan's or- 
ganization of the General Electric, United States Steel and Interna- 
tional Harvester companies, reflected the popular suspicion of Big 
Business. Almost continuously investigated by state legislatures, after 
its organization in 1870, the Standard Oil Corporation was already a 
national scapegoat by 1902, when Ida M. Tarbell's exposes of it began 
to appear in McClure's Magazine. 

Muckraking journalism increased the market for economic fiction. 
Novelists could represent business transactions with the assurance of 
both knowledgeable and partisan readers. Fiction and journalism were 
often scarcely distinguishable. Upton Sinclair insisted that The Jungle 
(1906), his novel about the meat-packing industry, had the validity of 
a sociological study. The economic novels of Robert Herrick, Frank 
Norris and Theodore Dreiser were just as stoutly documented. 

Publishers worked the appeal both ways. Charles Edward Russell, a 
New York reporter, published a factual report on the meat-packing 
industry, The Greatest Trust in the World (1905). It was a series of 
magazine articles documented by maps, tables and graphs, and repro- 
ductions of forged trade bills. But aware of readers' wishes, the pub- 

After Marquand, the Deluge 161 

lisher advertised the book as a "tragic romance of modern business 

The methods of newspaper reporting and a strong editorial legacy 
account for the distinctive qualities of American business novels. Their 
subject is always timely, and their tone is urgent. In both respects they 
have kept pace with the vast transformations in the business world 
since the end of the nineteenth century. New industries with new 
products for new markets have helped revolutionize financing, distribu- 
tion and selling, as well as manufacture. Through all these changes the 
economic novel has kept tabs on the businessman's response to busi- 

Business novels still worry over the fact that business swallows up 
the individual, but they have updated the subject. In the timely ver- 
sions of this plight the individual's impotence is inevitable. Being 
swallowed up, in fact, is the only way to succeed in business. Keeping 
up with the times, therefore, contemporary business novels are de- 
voted to the pathology of success. 

Mannerism begins to show up in the novels over the attitudes they 
urge toward this presumed inevitability of success. Popular fiction al- 
ways turns out to have gratified people's wishes in an apparently con- 
vincing way. Popular fiction in America has always been in the peculiar 
position of having to justify success. The business novel is a persistent, 
local instance of this. 

Success can be indiscriminately applauded, which is easy to do; it 
can be rationalized as inevitable and therefore right, which is a little 
more involved; or it can be explained, which is still more complex. 
Among contemporary business novels, the first which gained any popu- 
larity by trying to explain this inevitability was John Marquand's Point 
of No Return (1949). The present pathology of success in business 
novels really began with this book, which made a great many, and 
many different, audiences aware of a situation; and it made a singular 
impression on the book trade. In the flood of business novels which 
deluged the market after Point of No Return even the waves had 

After appearing in abridged form in the Atlantic and then as a five- 

1 62 The Denatured Novel 

part serial in Ladies Home Journal the novel was published in 1949. 
During the summer of that year the Book-of-the-Month Club offered it 
as a dividend. Grosset & Dunlap then published two editions of Point 
of No Return; and by the time Bantam's first paperbound edition ap- 
peared in 1952 well over a half-million copies had already been sold 
through trade, book-club, and reprint distribution. Later in 1952 the 
book was adapted to the stage and began a year's run on Broadway. 

The success of Marquand's business novel gave point to Alfred 
Knopfs remark (which he originally made in 1951) about buying a 
sure thing when you buy Marquand's books. Although not sensational, 
the book's sales were high, and they reflected the steady buying of dis- 
tinctly different markets. The fact that eight different publishers have 
since anthologized parts of the novel indicates its appeal to a variety 
of tastes and judgments. Also, it had created this appeal at a time of a 
rising market for novels in paperbound editions. Mannerism was one 
obvious result. Clusters of business novels thematically similar to 
Point of No Return began to appear in the early fifties. 

Point of No Return provided a point of departure for most of its 
successors, It dramatized a situation many novels have subsequently 
caricatured by exaggerating and simplifying it, But Marquand took a 
satirical not a popular view of his subject, and the mannerists, 
avoiding this view, have ended up with attitudes which do not fit the 

Marquand's novel dramatizes Charles Gray's career in the Stuyve- 
sant Bank to that climactic point when he learns whether he or Roger 
Blakesley is to become the bank's newest vice-president. The job will 
ultimately lead to the presidency, and each man has committed him- 
self and his family to a war of attrition for a prize which only one of 
them can have. It would seem so simple: Gray has an adversary com- 
peting with him for the favor of the bank's president, Tony Burton, 
yet this exhaustive personal conflict turns out to be irrelevant. 

Gray realizes he cannot reduce the bank to any manageable or satis- 
fying terms. "The Stuyvesant was the aggregate of the character of 
many individuals, who merged a part of their personal strivings and 
ambitions into a common effort." Gray conjures up metaphors to ex- 
plain the bank to himself. "It was like a head of living coral rising 

After Marquand, the Deluge 163 

above the surf, a small outcropping of a greater reef/' And all of its 
individuals, from the doorman to the directors, "were . . . asses follow- 
ing their bundles of hay." "They were all on an assembly line, but you 
could not blame the line. It was too cumbersome, too inhumanely 
human for anyone to blame." 

A few pages before the end of the novel, Gray is still trying to ap- 
praise it all. He and his wife have been invited to dine at Burton's 
home, obviously to be told the fate of his career. After a few more 
hours of attrition for the Grays, in social small talk with the boss and 
his wife, the gentlemen retire to the library, and Gray concentrates on 
the imminent disaster. In that short walk across the hall, "he was actu- 
ally walking . . . over the road of his career, a feeble little human track 
like the progress of a sea creature in the sand/' 

Having seen his own tiny image, Gray begins to adjust to it. His 
career is as good as over, and he must retreat to a new position. He and 
his wife will sell their suburban house and move into a smaller place; 
they will give up their plans for the more exclusive country club and 
for private schools. He begins to realize how tired he has been. He will 
never have to try so hard again, or be so cautious or obsequious. Then 
a deep sense of relief overcomes him. " 'It's over/ he said to himself as 
he walked across the hall. 'Thank God, it's over/ It was the first time 
he had felt really free." But the shocking news of success explodes this 

Gray suddenly learns that he has won the job, in a manner that stuns 
him. Burton casually announces the fact, then shows surprise that there 
should even be any doubt. There has been no conflict between Gray 
and Blakesley as far as the Bank is concerned. "You never thought any 
of us were considering Blakesley seriously, did you?" Blakesley was just 
not the right material; and as for the vice-presidency, says Burton, "it 
never occurred to me that you'd have any doubts about it." 

It paralyzes Gray to learn that he has never had any control of the 
situation, that he has no more power to refuse the honor than he had 
in winning it. He feels dull and very tired. He foresees a new profes- 
sional friendship with his employer, and transactions for a larger house, 
a new car, a more exclusive country club. Even before the end of the 
conversation he realizes he has never left the treadmill. 

164 The Denatured Novel 

Gray has made no decision. His success, already decided for him, is 
analogous to his entire life. He cannot even discover where the point 
of no return occurred, There has been no point at all, but a line stretch- 
ing back into his past, across which he could never step. 

Growing up in a small New England village, Charles Gray has al- 
ways known the primitive rituals of the caste system. But "system" 
means more in this book than merely social stratification. To Gray's 
father, for instance, it has meant the arrangement of one's whole eco- 
nomic existence. "The system is not fluid and it is very hard to beat," 
he says. "What system?" Charles asks. "Why, the system under which 
we live. . . . The order. There's always some sort of order." 

The hero's father has cultivated an independence in social matters 
which has made him an eccentric to the rest of the community, as if 
he were trying to beat the system by scorning it; then, following the 
loss of an inherited fortune on the stock market, he proves his point 
by suicide. The son absorbs the lesson; as he later explains to his wife: 
"There's no use getting mad at the system. We're part of a system 
where there's always someone waiting to kick you in the teeth in a 
nice way." Gray solves the system by acceding to it, and the plot of the 
novel bears him out. 

Charles Gray's debilitating success dramatizes the peculiar course of 
literary naturalism in America. You can see how small the businessman 
has shrunk when you measure Gray by another hero, Silas Lapham, 
about whom William Dean Howells wrote, sixty-four years earlier. The 
comparison is useful. The Rise of Silas Lapham ( 1885 ) presents a hero 
with a moral problem, involving financial success or failure, which he 
resolves by himself. Howells idealizes the individual's control of affairs, 
and thereby hangs another tale. 

Silas Lapham manufactures a paint whose excellence has made him 
wealthy, and he determines to use this wealth to improve his family's 
social position. But Lapham has made a costly compromise with his 
conscience, having forced out of partnership the man whose money he 
once needed. When this partner returns and asks for help, LaphanVs 
wife prevails on her husband to pay off his conscience. The ex-partner 
is dishonest, but Lapham pointedly ignores the bad risk and loans him 
large amounts just before his paint business begins to fail. 

After Marquand, the Deluge 165 

The hero must capitalize his business or go bankrupt, but he has 
loaned his capital and spent his credit. At this point the ex-partner 
reappears, having found some purchasers for the land Lapham holds as 
collateral on the loans, and which both men know to be worthless. The 
hero's dilemma is clear: to sell the land and recover his loss, or to refuse 
to sell and save the other party, and the author subjects him to all the 
tortures of righteousness. 

Lapham spends the night pacing his room, and Howells likens his 
struggle to Jacob wrestling with the angel. Like Jacob, following the 
struggle, Lapham finds peace. He has victoriously decided to go bank- 
rupt. Virtue triumphs. He loses the battle to win the great moral war. 

The picture of this Victorian businessman wrestling with the angel 
apparently gratified the women who read the novel and dreamed of 
what a virtuous world it would be if their husbands would only listen 
to them. The fun of justifying success consisted in deciding the terms 
on which one would accept it. But what made the dream convincing 
was the assumption, which Howells shared with his readers, that man 
was a free agent in the first place. 

Marquand and his readers share no such luxury. Within the literary 
history which contains both novels the fact of "scientific" determinism 
separates them. Toward the end of the nineteenth century, this doc- 
trine became incorporated in a theory of composition known as literary 
naturalism, which aimed to document determinism in copious and 
authentic detail. With the illusion of complete objectivity, according 
to this theory, the novelist would prove that the individual man is a 
small and limited creature, determined by heredity or by the social or 
economic forces of his environment. 

No one has yet written the novel which could perfectly conform to 
Emile Zola's definition of this theory not even Zola. Because natural- 
ism denies any special significance to human beings, it therefore denies 
the vitality of fiction. Nevertheless, the prestige of literary naturalism 
in Germany and France, the translation of Zola's literary criticism into 
English, and American fiction's affair with journalism have so influ- 
enced writers that virtually every economic novel in America after 1900 
at least implies some reference to naturalism. 

This theory reflects the change in the businessman from an active 

1 66 The Denatured Novel 

hero to a passive one, from a Silas Lapham to a Charles Gray, Business 
history records substantial cause for the extinction of a Silas Lapham 
from the business world. Lapham was an entrepreneur, owner and 
manager. During the twentieth century, however, business organiza- 
tion outgrew his kind. With the development of the great public cor- 
porations, even before World War I, ownership and control began to 
separate and to offer disparate images to the public mind. 

During the thirties the concept of ownership changed as Big Busi- 
ness strenuously publicized the notion of a people's capitalism, de- 
pendent on many small stockholders. As for control, people gradually 
heard about a new kind of businessman, called the administrator. Here 
was a new image of the businessman. The image was anticipated by 
Adolphe A. Berle and Gardiner C. Means in their epochal study, The 
Modern Corporation and Private Property (1932). They discovered 
that some two hundred non-banking corporations dominated the na- 
tional economy and that the control of these corporations lay not in 
their legal owners, the stockholders, but in their executive managers. 

Manuals of business management sold briskly as early as the igzo's. 
By the time of World War II, James Burnham's The Managerial Rev- 
olution (1941) dramatized the new apotheosis of business adminis- 
trators. The nation, said Burnham, must now acknowledge business 
managers as a new social group. This new professional manager actu- 
ally controlled the instruments of production. But even the manager 
was only an employee. He had no function by himself. He was only a 
unit of the great business society. 

No wonder Charles Gray sees himself as a small sea animal and the 
Stuyvesant Bank as the outcropping of a reef. Point of No Return 
uniquely represents its times, as Howells' novel once did. But in Mar- 
quand's view Charles Gray's promotion to the Bank's vice-presidency 
is distressing. This is where Marquand's successors part company with 
him. But they don't part very straight. 

These mannerists exaggerate the plight of the businessman over- 
come by success, but they avoid any honest dilemma over it. Instead, 
they celebrate it. Their main problem, therefore, is to make it look as 
though there were a problem in the first place between the business- 
man and this benevolent business system. The best way to see this 

After Marquand, the Deluge i by 

mannerism at work is to compare some of the caricatures of the passive 
businessman which followed Point of No Return to market. 

Between 1952 and 1956, the five years during which the first paper- 
bound edition of Marquand's novel remained in print, the most com- 
mercially successful business novels were Cameron Hawley's Executive 
Suite (1952) and CashMcCall (1954), Howard Swiggett's The Power 
and the Prize (1954), and Sloan Wilson's The Man in the Gray Flan- 
nel Suit (1955). All of these books appeared in paperbound editions, 
two of them had additional book-club sales, and all of them were 
adapted to motion pictures. Each earned more for its author than John 
Marquand's novel had earned. 

These books variously emphasize the most recognizable qualities of 
Point of No Return. A fifth novel, published during this period, is 
George DeMare's The Empire and Martin Brill (1956). This novel 
amplifies the qualities of Point of No Return as well as those in its 
four other predecessors. But all of these latter novels are eclectic; they 
are all stylized. 

The narratives differ. In all of them, however, a particular business 
enterprise becomes both the emblem and index of "the system." And 
most of them have inflated the net assets and size of the business and 
its control over the individual. The change from "company" to "cor- 
poration" and then to "empire" suggests their magnification of the 
system. Even Howard Swiggett's The Power and the Prize, the most 
apparently humanistic of them all, insinuates a benevolent determin- 
ism by a board of directors. 

Swiggett's novel tells about Cleves Barwick, the executive officer of 
Allied Metals Corporation, -and his endeavor to form an international 
company that will revolutionize the world production of nonferrous 
metals. But Barwick falls in love with a woman of whom the chairman 
of the board cannot approve. A refugee from Central Europe and a 
suspected Communist sympathizer, she would presumably become a 
liability to the Corporation. The chairman discredits Barwick's choice 
in such a way as to jeopardize the negotiations for the new company. 
Clearly, Barwick must choose between the woman and the company. 

His defense of his fiancee rests solely on her denial of having been a 
Communist, and apparently he can reconcile the conflict only by lying. 

1 68 The Denatured Novel 

The Corporation's directors give him a chance to declare that he has 
already investigated her political past, thereby clearing the Corpora- 
tion of any guilt which may befall her. Barwick refuses to lie, but an 
old movie "gimmick" saves him. The directors were merely testing his 
integrity. They know he has not investigated his fiancee, so instead of 
firing him for his testimony they promote him to the chairmanship of 
the board. With this endorsement he completes negotiations for the 
new company before embarking on his honeymoon. 

Here is the old-fashioned conflict between love and duty, but Swig- 
gett has updated it. The hero with a choice to make is not really a free 
agent. Although Barwick has maintained his right not to explain any- 
thing, the circumstances has been rigged. He is on trial before a jury 
which already possesses the facts and which conducts the prosecution. 
At the moment of his triumph Barwick is less the protagonist than he 
appeared. As with Charles Gray the system has acted benevolently, but 
in this novel nobody worries about it. 

One might miss this subtle downgrading of the hero in an exalted 
system had not Howard Swiggett literally repeated himself in his next 
and final novel, The Durable Fire (1957). In this latter version of the 
same story the hero argues with his uninformed and hypocritical boss, 
and the case goes before the directors of the Corporation. It turns out, 
however, that these representatives of the benevolent system already 
know the facts and merely wish the hero to justify their decision in his 

Business enterprise even more thoroughly pre-empts the individual 
in Cameron Hawley's Cash McCall; and by another round of inflation 
the presiding force of this business world appears even further removed 
from the company employees. This Olympian force is itself a business: 
the purchase and sale of corporations, sustained by the federal tax 
law's provision for capital gains. In this new world the oldtime religion 
of company loyalty loses favor, and the novel's apparent conflict pro- 
ceeds from this fact. 

Ownership has become a commodity. Brokers manipulate compa- 
nies, often fraudulently for profit, denying the permanent value of a 
business and its service to the community. So it seems to Gil Clark, a 
young man dedicated to managing the Suffolk Moulding Company, 

After Marquand, the Deluge 1 69 

which will apparently be sacrificed to this sort of manipulation. Gil 
finally meets his adversary, Cash McCall, the symbol of this system, 
the manipulator whose name has become legend. 

But this Croesus confounds all expectations. He operates legally. "I 
don't make the rules, Gil. I only play the game. I never thought much 
of making the kick-for-point after touchdown, either, but as long as 
it's in the rule book, that's the way the game is played." What could 
be more American? Or more unassailably moral? "There's only one way 
I can get a wallop out of a deal like this, Gil. And that's by way of 
knowing that I haven't dug money out of another man's hide." 

McCall is downright benign, as any magazine reader could tell in an 
instant: he is in his "late thirties, possibly forty . . . more like a pro- 
fessional athlete than a businessman." He also has impeccable man- 
ners, and although he owns unbelievable wealth (he smiles with a 
"purse-string puckering around the eyes") the name "Cash," rumors 
notwithstanding, he has legitimately inherited from his mother's fam- 
ily. Some more of his heritage conveniently explains his motives: 
through his father's futile wage-slavery Cash has learned to disdain 
company worship. 

Having given him necessary credentials, the author enhances the 
hero with a kind of parental authority. He owns most of the corpora- 
tions in sight, including the one which has currently employed Gil 
Clark. McCall has, in fact, secretly maneuvered that young man's 
career into his own orbit. As for Suffolk Moulding, the company whose 
disposition he presently controls, Cash McCall has only the most re- 
sponsible intention. It turns out that young Gil Clark did not have any 
problem at all. He only thought he did. 

McCall merges Suffolk Moulding with two other properties he has 
dramatically acquired, saving the stockholders including a medical 
research foundation from ruin, and so demonstrates his superiority 
to lesser-minded company men. His supra-business really nurtures 
what it seems to prey on. The benign force comfortably ruling the 
economic world, which depressed John Marquand's hero, has here 
become a happy inevitability. 

This novel celebrates earned income large amounts of it for its 
own sake. Cash McCall even delivers a lay sermon, scolding Americans 

170 The Denatured Novel 

for their public discomfort over wealth. But faced with precisely this 
discomfort in the public mind and wishing to justify profit, the novel- 
ist trafficks in attitudes. The sanctioning doctrine of Good Works has 
been at hand ever since Benjamin Franklin canonized the Arminian 
Heresy. So Cash McCall, like his patron saint, does well by doing good. 

The happy doctrine of profit through Good Works also explains 
Cameron Hawley's earlier novel, Executive Suite, a splendid suspense 
story with a little something for everyone. This book deifies the mate- 
rialist, but so covertly that the reader is never offended by his own 
appetite. The sudden death of Avery Bullard precipitates a crisis in the 
furniture business. Which of the five vice-presidents of the Corpora- 
tion will succeed the late Mr. Bullard to the presidency? The leading 
contender is the controller, Loren Shaw, a relentless and aggressive 
man dedicated to company profits. 

Shaw is a money grubber, a conniving materialist, and the book 
makes a heavy case against him, casting suspicion on all his motives. 
The movement in the company to stop Lorcn Shaw quickly settles 
on MacDonald Walling, the youngest of the vice-presidents. The ero- 
sive battle between these two recalls the earlier, mannered contest be- 
tween Charles Gray and the man he thought he had to beat. In this 
one, however, Don Walling triumphs much more theatrically. He wins 
the directors' votes at a show-down with Shaw by an impassioned 
speech which states the theme of the novel or seems to. 

Walling's convictions are not at first entirely clear. He damns Mam- 
mon, specifically Loren Shaw's priority of stock dividends, he applauds 
high-quality products and recognizes the employees as human, and he 
promises new and greater growth for the Company. These comforting 
attitudes seem safe enough. But the rhetoric of Mr. Walling disguises 
some fortuitous illogic. Speaking in anger, he excoriates Avery Bullard, 
who had had an eye on the profits, who had "been so busy building a 
great production machine that he ... lost sight of why he was building 
it/' Yet Walling earns the vote, after some talk of making newer and 
better furniture, by promising to build an even bigger company. 

This young man surely offers the directors the image of his predeces- 
sor. But at the time the author throws the switch, after Walling's 
rhetoric, the directors scarcely realize that a vote for Walling is a vote 

After Marquand, the Deluge 171 

for Shaw. This fact becomes magnificently clear when the visionary 
Walling appoints Shaw, the money-man, to the new executive vice- 
presidency, explaining: "I'll need somebody to help me keep my feet 
on the ground." 

Further evidence of this union of purpose presumably too complex 
never reached the movies. In the film Loren Shaw bitterly concedes 
defeat, but not in the novel. Perceiving that Walling's presidency will 
be the best for Tredway and for its dividends, Shaw is among the first 
of the voters to throw his support to the young man. All's well with 
the company. The new president has sold a bill of goods. Unlike his 
adversary he has managed to present his materialism not merely as a 
matter of profit but as a mission to be celebrated. What counts with 
the directors is the moral hocus-pocus which disguises the materialism 
in the first place. 

Slick fiction, John Marquand's included, contrives to make it easy 
for the hero to take what is coming to him. What comes is material 
well-being. But these later novels differ from their predecessor by re- 
nouncing the uncomfortable sense of compromise and by presenting 
instead some justification for the hero's happy retribution. The most 
striking example of such expedience is Sloan Wilson's The Man in 
the Gray Flannel Suit. 

Like Point of No Return this tale begins in authenticity, telling the 
problems of a pleasant young man with his attractive wife and their 
average children, living in a normal, hectic world. Tom Rath has two 
problems, professional and private, which equally urge him on to self- 
discovery. Professionally, he must decide whether or not to be a busi- 
nessman to commit himself to executive endeavors in the communi- 
cations industry or to find a less demanding job which will offer dig- 
nity and spare time. Privately, he must decide whether or not to tell 
his wife about his paternity of an illegitimate child in Italy during the 
war. In the manner of a morality tale, each problem offers the hero 
neat and exclusive alternatives. 

And he decides: professionally, to abandon the chance to be a gladi- 
ator and to accept the lesser job; privately, to tell his wife about his 
illegitimate son, and to secure her blessings in settling an annuity upon 
the lad. But as it turns out, the hero has had no problem at all, because 

172 The Denatured Novel 

he could afford, in each case, to make the morally preferable decision. 
In the novel's most revealing sentence, Tom Rath perceives that 
"money is the root of all order." Since he has inherited his grand- 
mother's estate before having to decide his problems, his choices have 
none of the moral significance proclaimed for them. The hero has 
merely had to wait for circumstances to make it easy. 

In this parody of Virtue Rewarded the reward comes first, and the 
author's contrivance makes a caricature of the benevolent society. 
When Tom Rath's legacy appears doubtful because of a counter-claim, 
for instance, a probate judge goes out of his way to investigate the other 
claimant and to discover his dishonesty. The villagers, in effect, en- 
dorse a drastic exception to the zoning laws, allowing Tom the oppor- 
tuity to turn his inherited land into a profitable development. Finally, 
when the hero denies the business opportunity offered him, even 
chastising his boss in the process, that unbelievably tolerant benefactor 
obtains for the hero a life-time sinecure. 

It is all so appallingly easy. The novel scarcely honors its publisher's 
claim that it speaks for a whole generation. On the contrary, this hero's 
endowments are so special and his society's dispensations so tailored to 
his needs, that his gray flannel suit would not fit anyone else. 

Stylization exaggerates gestures and attitudes. Each of these novels 
offers a comparative, and some a superlative, to what John Marquand 
has set down. Competition for the top job is more erosive, more ob- 
viously dramatic in Executive Suite; the hero is more willfully passive 
in Sloan Wilson's novel; and the company more specifically paternal 
in The Power and the Prize, yet dwarfed in Cash McCall by an eco- 
nomic force above and beyond even the corporation. In short, whatever 
the problem, it is more urgent in these eclectic novels. 

These stylized attitudes are rendered by stylized conventions. The 
author's chore is to dispose satisfactorily of the hero's problem; or the 
other way around, to make it seem for awhile as though the hero has a 
problem. It is the same either way. Some evidence need only be planted 
early in the story, which the climax can then recall in reversing the sit- 
uation. This is the "gimmick" in the magazines and the movies. 

John Marquand plants his evidence for the climax of Charles Gray's 
struggle in a long, detailed flashback. This is Marquand's most promi- 

After Marquand, the Deluge 173 

nent convention. The hero's past is the majority of the novel, and the 
whole of it is the evidence for the way things suddenly turn out. These 
latter novels by the other authors, however, stylize the effect of sudden 
reversal by leaving out the past which precedes it. The climax is more 
sensational that way, less mundane and more dream-like. 

In Point of No Return the hero's success is decided for him because 
this is the way it has been all his life. The sequence of events accounts 
for this apparently incontestable logic. Marquand introduces the prob- 
lem in the narrative present, then suspends it for a thorough and pains- 
taking exploration of Charles Gray's past, during which the hero is 
seen time and again abiding by accepted conventions of behavior. 

Growing up in a small town, Gray conducts a courtship across the 
tracks and then dutifully abandons it. He conforms to the accepted 
image of a young man in a brokerage house, even playing the stock 
market conservatively in an era of wild speculation. He repeatedly con- 
templates the joys of unconventional behavior, yet keeps his thoughts 
to himself. Tempering his private observations with public good sense, 
he earns the complaisance and benediction of friends and employers. 
This pattern repeats itself so often, throughout the middle half of the 
novel, that when Marquand resumes the problem of the vice-presi- 
dency, he has already and repeatedly signaled its resolution. 

The narrative past has determined the narrative present. Charles 
Gray's conservative honesty in deference to the system is bound to win 
the system's approval. Once you know the past, granting a consistent 
character, there is no present problem. What at first appeared prob- 
lematic turns out to be merely a foregone conclusion. So it is with 
H. M. Pulham and George Apley, in fact with most of Marquand's 

Marquand manages a convincing surprise by making the evidence 
so obvious that the reader takes it for granted. The problem has appar- 
ently solved itself; it is a kind of optical illusion. This surprise by illu- 
sion is precisely the quality in Marquand's narrative method which 
succeeding novels have exaggerated. By a kind of fictional shorthand 
they have abridged the past tense, which in Marquand's novels solves 
the present problem. 

None of these latter heroes has a past in the sense of a history al- 

1 74 * lie LJenaiurea i\ ova 

ready resolved. Except for fragmentary flashbacks identifying the char- 
acters, these novels remain continuously in the present. They have 
abridged that part of the narrative which accomplishes the solution, 
but they must still fulfill the fictional requirement of a conflict solved, 
They must come out even at the end. So instead of solving the problem 
they dissolve it. 

This tour dc force is like the solution to a puzzle, and just about as 
significant. Sloan Wilson's hero did not have to work at a distasteful 
job in order to support two families; he only thought he did, until his 
inheritance made everything dreadfully easy. Gil Clark, in Cameron 
Hawlcy's novel, merely thought the brokerage of corporations was 
malevolent, until Cash McCall revealed the truth. Don Walling never 
challenged the primacy of company dividends, in Executive Suite; it 
just took a little time for him to reveal himself as a more imaginative 
money-man than his adversary. And was there ever any doubt that the 
directors would endorse Clcve Barwick's high-principled courtship? 
Only by assertion. 

The difference between John Marquand's patently logical solution 
of a problem and the mere illusion of the problem in the first place 
also makes the difference between an attitude and the commercial ex- 
ploitation of it. These latter books all say less than they seem to. By 
dismissing the problems they have posed they disqualify the attitudes 
these problems seem to have produced. The greater the exaggeration 
in the first place, therefore, the more obvious the fake. In the extreme 
this affliction reduces a novel to saying nothing of its own. 

The fifth and most recent in this cluster of business novels, George 
DeMare's The Empire and Martin Brill, reaches this point of ob- 
livion. It tells of five employees in the public-relations department of a 
vast and anonymous organization. This nameless corporation, which 
sells an unknown product, is called merely "the Company," "the Sys- 
tem," or "the Empire." 

During World War II, the time of the story, the Empire's public- 
relations problem is to broadcast its vital importance to the national 
war effort. Its headquarters occupy a skyscraper in a large Eastern city, 
and the nerve center of these headquarters is on the top floor. The 
reader catches a glimpse of two vice-presidents, one floor below, mys- 

After Marquand, the Deluge 175 

teriously referring to the "Old Man/' but that is as close as anyone gets 
to the soul of this system. According to the blurb on the paperbound 
edition ( 1957), "it might be called the story of the successful 'company 
man.' But it is also the story of hundreds of thousands of 'big company' 
employees whose dreams are never fulfilled, who find themselves im- 
prisoned in a treadmill of frustration/' 

Of the novel's five case histories, one man commits suicide because 
his mistress denies him. The second becomes an alcoholic, loses his 
job, and tramps the streets. The careers of numbers three and four con- 
tradict each other. Number three stands firm on a moral matter (like 
Cleves Barwick in The Power and The Prize), gets promoted when 
events turn out to justify his position, and passes from view to the upper 
stories of the skyscraper. 

Number four, who is morally dishonest, seeks preferment at the ex- 
pense of his colleagues. When his connivance is discovered, the Empire 
awards him a horizontal promotion, and he passes from view among 
the lower stories of the skyscraper. The fifth man, Martin Brill, is the 
hero only by deference of the novel's title. He is kicked around the 
Empire and finally promoted to the chair of the man who first hired 
him. But on the way to this promotion he has suffered battle fatigue. 

This schematization of the five characters accommodates the anony- 
mity of the Empire. The only ruling force in the System is the irony of 
denial. The successful executive kills himself, the dedicated company- 
man gets fired. The employee who renounces preferment wins, and the 
one who puts it before all else loses. Martin Brill, the man alleged to 
have great expectations in the Empire, gets unaccountably shuffled 
off into a minor job and exiled from headquarters. Then, having given 
up, he unaccountably wins. 

The anonymity, the irony, the lack of any cipherable plan, and the 
boss nobody ever sees arc all symptoms of literary naturalism. More- 
over, signs of earlier naturalists mark the final impression of this book. 

There in the misty afterglow of the fading light, it towered like a huge 
and shadowy fortress, heavy and sinister, rising out of the sea. It was 
the Building the great, shadowy Building. It seemed to him, as he 

1 j6 The Denatured Novel 

stood there gazing at it from the distance, as if it 'were a symbol of the 
whole vast Empire the granite stronghold of the mysterious powers 
of the age fabulous, immense, and strange. 

If you recall the static symbol of the Chicago Board of Trade Building, 
with which Frank Norris concluded The Pit (1903), you might marvel 
at the coincidence. 

But this eclectic novel has also absorbed and exaggerated more re- 
cent determinism. It capitalizes the "System" and in great detail in- 
flates the concept of its benignity. "It was only an impersonal, cold 
conglomeration of hundreds of thousands of unsentimental, common- 
place people, machines and buildings, tracks and towers, lights and 
voices; but somehow it was a solid thing, and it was there to stay." The 
individuals within it are as nothing, 

. . . but the Company which would see thousands like them come and 
go, rise and fall, would endure, would remain there forever, substantial, 
unchanging, powerful, enjoying a species of immortality, not as hu- 
mans but as humanity itself, a monument to the creative force and en- 
durance of collective men, no matter how mediocre and to the power 
of human need. 

Marquand's metaphor of the Stuyvesant Bank, the "head of living 
coral rising above the surf, a small outcropping of a greater reef," offers 
a comparable realization, but it is simpler and more dynamic. The 
comparison between one man's metaphor and another man's catalog 
suggests the entire relationship between these two novels. 

Perhaps there were more salable attitudes for this latter novel to ab- 
sorb. The "system" in it is more pervasive than in Point of No Return 
and more impersonal than in Cash McCall. The retributions of the 
Empire recall the rewards and penalties of the board of directors in 
The Power and The Prize, although retribution no longer has any sense 
of necessity, The Empire and Martin Brill attempts to show the 
modus operandi of the big company. But the anonymity defeats this, 

After Marquand, the Deluge 177 

and the machinations degenerate instead into office politics, like those 
in Executive Suite. 

The Empire and Martin Brill is no copy. It is an original artifact. 
The author's personal experience in just such an actual business cor- 
poration, in fact, inspired his convictions. But he represented these con- 
victions by literary conventions which served other needs in other 
novels. His compendium consequently attracts no central or com- 
pelling idea to his particular subject. Reading it is like wandering 
through a hall of mirrors, seeing only the distorted reflection of other 

This particular cluster of economic novels is a small symptom of 
a general condition in the book trade. The fiction of business has got 
thoroughly mixed up with the business of fiction. Most other "profes- 
sional" novels, as they are called about doctors or lawyers or police- 
men, about Madison Avenue or Hollywood tell the same story. The 
overproduction of fiction follows the elemental fact that a successful 
book attracts more like it. Each new stylization is more like it than 
the last. 

There is no end to this subject, but what happens to the novel form 
must be obvious. It deteriorates until some writer conceives of a novel 
out of his own experience and organizes the actual chaos of things into 
an ordered, complex whole which appraises the way of the world. He is 
an artist. If he can write well enough and if he has unaccountable good 
luck, people will talk about his book and even buy it. 

On the strength of this, publishers will spot a "trend." Then the 
whole dreary process will start again. A cluster of new novels will be- 
gin to appear: packaged, advertised, some of them even written to ex- 
ploit and exaggerate the characteristics of the original book, but always 
contrived to attenuate it- This is all so terribly calculable, except for the 
artist in the first place, thank God. 

Chapter VIII 

Arms and the Manner 

THE human ideal of goodness has always had a hard time of it, which 
explains why the triumph of virtue makes the headlines and why 
fiction is so necessary. Conflict is a constant in fiction, but its local 
metaphor changes, keeping up with the times. By mid-century in 
America the most persistent local metaphor of the novel's conflict was 
warfare for an obvious reason. So far in the twentieth century warfare 
has been the most compelling denial of the idea of human efficacy in 
God's world. 

Contemporary book publishing has reflected this notoriety. In 1948 
Publishers Weekly counted some 270 American novels whose subjects 
involved World War II, many of them published before the war had 
ended. While the total number of published novels was steadily 
shrinking after V-J Day, the number of war novels began to increase; 
1948 was a crucial year for war books. The large sales of Dwight Eisen- 
hower's Crusade in Europe, Winston Churchill's The Gathering 
Storm, and the early volumes of official military histories reflected 
widespread interest. The fiction market throve on Norman Mailer's 
sensational first novel, The Naked and The Dead. 

Mailer's novel profited by an extraordinary advertising campaign. 
A series of teaser ads in trade and review journals stimulated interest 
many months before publication. Advance sales were high, and the 
novel continued to sell: 137,000 copies through the trade in eight 
months. The Book-of-the-Month Club distributed 100,000 copies and 
the Book Find Club another 64,000; and still the trade sales prospered, 

Arms and the Manner 179 

totaling over 200,000 by January 1951, when the novel appeared in a 
papcrbound edition. 

Later in 1948 Irwin Shaw's novel, The Young Lions, made almost 
as much of an impression. It was distributed by the Fiction Book Club, 
and the Post Hall Syndicate and Omnibook both purchased second- 
serial rights. The significant fact in the sales of both novels was their 
stamina in the market. It was three years before The Naked and the 
Dead went into a paperback edition, and during this time 200,000 
copies of the trade edition were sold. The sales of each novel in paper- 
back now total more than 1 5,000,000 copies, and are still increasing. 

In 1951, a few months after the publication of the paperbound edi- 
tion of The Naked and the Dead, two other war novels appeared, with 
even more conspicuous success. The Caine Mutiny (1951; 1954), by 
Herman Wouk, sold over 200,000 copies through the trade in its first 
nine months, and by the end of 1952 it had nearly doubled that total. 
Its book-club distribution (by Book-of-the-Month Club, Literary 
Guild, Peoples' Book Club, and Doubleday Dollar Book Club) was 
unprecedented. The Caine Mutiny was also condensed for the Reader's 
Digest Book Club, serialized in some forty-five newspapers, and 
adapted to the stage for both Broadway and road-company produc- 
tions; and like Shaw's and Mailer's novels it was filmed in the grand 
manner. The other novel of 1951, James Jones's From Here to Eternity, 
has sold more than 2,500,000 copies in reprint form. It was the first of 
these four novels to be filmed, and on the strength of its film it has 
achieved the largest sale of all of them in the shortest time. 

For nearly ten years, beginning with the publication of Mailer's 
novel, the new market steadily increased. There were all sorts of rea- 
sons for it. Millions of Americans home from the war wanted to read 
about themselves. The crisis of Berlin and the American airlift, and 
then the police action in Korea made warfare imminent again. Even 
more imminent was the film industry's capitulation to television, until 
Hollywood successfully gambled on the wide screen in 1954. This last- 
gasp invention created an urgent need for spectacular story material. 
The war novel satisfied all of these interests. 

One dissonant note occurred in this chorus of war novels, but at the 
time it was not very loud. Ira Wolfert's An Act of Love (1948; 1955) 

1 8o The Denatured Novel 

had been published along with The Naked and the Dead and The 
Young Lions. It was promoted as a "Big" war novel like the others, but 
it did not share in their riches. The success of the other books had ap- 
parently no effect on its sales. Ten times as many copies of Mailer's 
novel were sold in eight months as were ever sold of Wolfert's novel. 
Wolfert's trouble was that he saw the war as a problem in ethics by be- 
lieving in the efficacy of his hero; and the relatively poor showing of 
his book was prophetic. 

But the four famous novels had tremendous effect. Even before 
Hollywood's wide screen raised the profit in war fiction, the influence 
of these four books became apparent on the lists of reprint publishers. 
Reprinters began purchasing the rights to novels published before the 
campaign of '48 and still in trade editions. The publishing dates of the 
trade and reprint editions of Frederic Wakeman's Shore Leave (1944; 

1948) represent a familiar pattern, and the pattern was still evident ten 
years later with the belated reprinting of Lawrence Kahn's Able One 
Four (1952) as The Tank Destroyers (1958) . These two reprints were 
not among the very few which rescued novels worth saving, such as 
Harry Brown's A Walk in the Sun ( 1944; 1957) or The Gallery ( 1947; 
1950), by John Home Burns. 

The search for newly salable material reached even to the fiction 
of World War I: Ernest Hemingway's A Farewell to Arms (1929; 

1949) for instance, Soldier's Pay (1926; 1951) by William Faulkner, 
William March's Company K (1933; 1958), and Eric Remarque's 
All Quiet on the Western Front (1929; 1958). By suggesting to book 
buyers a continuing tradition of war fiction these reissues of old titles 
made useful advertising copy for the newer fiction of World War II 
and the Korean conflict. Thus, the paperbound package of a Korean- 
War novel by J. D. Hollands entitled Able Company (1956; 1958) ex- 
plains in reviewers' jargon, that this " 'Great War Novel' . . . should 
rank with Remarque's All Quiet on the Western Front" a claim which 
stretches a coincidence of subject matter beyond all reason. 

The claim made for Hollands' novel is ludicrous. The only similarity 
between the two books lies in the kind of war each represents. The war 
in Korea, as in France more than thirty years before, was a war of posi- 
tion, and the ground fighting in both became an erosive stalemate. But 

Arms and the Manner 181 

two novels about war could scarcely differ more than these do. Re- 
marque's novel is selective and suggestive; Hollands' novel is a ful- 
some encyclopedia of dreary details. This difference, however, did not 
keep Hollands' reprinter from claiming fame by association. 

This is a typical trick, especially when one reprinter has published 
both of the novels on which the claim is based. The reprinter proclaims 
Lionel Shapiro's The Sixth of June (1955; 1956) to be "The greatest 
story of love and war since Hemingway's A Farewell to Arms" 
which he also reprinted. Sometimes this little device is even laughable. 
According to its paperback cover, Kenneth Dodson's Away All Boats 
(1954; 1 95^) i s "The mightiest novel of men at war since Battle Cry" 
which was originally published the same year. How could false infla- 
tion be more modest than this? 

Inflation is customary in this sort of claimed esteem. Battle Cry 
(1953; 1954), by Leon Uris, "shares the greatness of From Here to 
Eternity and The Caine Mutiny." Wirt Williams' novel, The Enemy 
(1951; 1956), "Takes its place . . . beside The Caine Mutiny, and the 
stories of Conrad and O'Neill." These likenesses belong to the advertis- 
ing, not the book. 

War and Peace is claimed among the credentials of The Sixth of 
June, which is about like comparing The Origin of the Species with 
Peter Rabbit. The borrowed esteem on the paper cover of Ross Carter's 
actual account of paratroopers in Europe, Those Devils in Baggy Pants 
(1951; 1952) "Not since The Naked and the Dead Mr. Roberts 
or South Pacific has there been a book like this!" patently denies 
the facts. The cover-claim of Lonnie Coleman's Ship's Company 
(1955) blandly contradicts itself: "Action, danger and high passion in 
a book that ranks with Mister Roberts." It is the old promotional de- 
vice of trying to cash in on someone else's best seller. These false 
claims might be dismissed were they not symptoms of the reason so 
many books are published in the first place. 

When the reprinter informs the customer that Robert Bowen's 
Bamboo (1953; ^55) is "Compared to From Here to Eternity" he is 
misrepresenting a literary idea. Bamboo is a short novel about a thirty- 
day leave which a regular-Navy sailor spends in the shacks and bars of 
Manila: drinking and brawling with his shipmates, living with a Fili- 

1 8z The Denatured Novel 

pino girl, and repenting his drunken brutalities. Glandular functions 
and vague remorse entirely comprise this character, of whom the au- 
thor makes no consequence. The only grounds for comparison with 
Jones's novel are the excesses of a thirty-year man on peacetime duty 
in the Pacific. Excess has some significance in Jones's novel, in the 
claustrophobic lives of a dozen characters, but it leads to nothing in 
Bamboo except the pale recollection of From Here to Eternity. 

Mannerism in war fiction extends beyond the advertising to the 
contents of the novels themselves. Most of the novels published after 
those four which built the current market are startlingly similar to one 
another, but they are similar in ways which their publishers would 
rather not mention. Like all other books which swarm after a best 
seller, these war novels exaggerate conventions which their predecessors 
developed. As a result the book trade recognizes a conglomerate 
product, published from time to time, which could be called the "nor- 
mal" novel. 

For the past ten or twelve years the normal war novel has been the 
most familiar species of contemporary American fiction. Its properties 
and the way they evolved, moreover, illustrate how the publishing in- 
dustry hastens and formalizes the degeneration of a species. This de- 
generation is similar to what has occurred in those clustered business 
novels whose authors followed John Marquand to market. But there 
are more factors involved in the degeneration of war novels. 

The "normal" war novel is partly the result of a widespread obses- 
sion for authenticity which the subject itself inspires, and partly the 
result of the stylizing of certain fictional conventions. This conglomer- 
ate product so obviously denies the possibilities of the novel form that 
it bears looking into. Its gross authenticity prevents fictional develop- 
ment and disqualifies the whole subject of moral choice; this can be 
seen first by observing the characteristics of some normal novels. Then 
it will be useful to notice the derivation of these characteristics, to in- 
spect the conventions of earlier war novels which the normal novels 
expediently distort. 

First, to the characteristics themselves. The normal novel is a "Big 
Book." It bulks large, it is heavily populated, and its subject is sen- 
sational. It involves timely attitudes about emotional disturbance, 

Arms and the Manner 183 

sexual difficulties and racial problems. Nearly everything about its sub- 
ject is excessive; even its manner is excessively ponderous. In the mis- 
taken attempt to be "brutally honest/' as the ads say, it is excessively 

The normal war novel is the saga of a group of private citizens assem- 
bled for military training and gradually built into a fighting unit, im- 
bued with camaraderie, and embarked for combat. When this group 
finally meets the enemy, the citizen-soldiers test themselves at terrible 
cost, and the whole erosive experience yields only a muted hope for the 
few survivors. Most of the soldier-writers have the same significant 
story to tell, with the same claim to authenticity and with the same 

The first narrator in William March's Company K, one of the re- 
vived novels about World War I, speaks for all soldier-authors. "I wish 
there were some way to take these stories and pin them to a huge 
wheel," he says, "each story hung on a different peg until the circle was 
completed. Then I would like to spin the wheel, faster and faster, until 
the things of which I have written took life and were re-created, and 
became part of the wheel, flowing toward each other, into each other; 
blurring, and then blending together into a composite whole, an un- 
ending circle of pain. . . ." He is obsessed with making a picture and a 
sound of war which encompasses everything else. 

This is the way it was, the author says. He was there and cannot be 
contradicted. Both the strength and the failure of the normal war 
novel proceed from this fact. The military units and the campaigns 
which involved them are matters of public knowledge and give a cer- 
tain signal value to the book. But the disadvantages of authenticity 
are crippling. The immediacy of the author's subject can overwhelm 
him, to the detriment of fiction. When the subject of war the denial 
of human efficacy overwhelms the author, his novel has a hard time 
convincing anyone. He can write only of determinism, and this contra- 
dicts the nature of the novel. 

The novel is a humanistic form. By simplifying chaos into conflict 
it poses alternatives for its characters; it presents the possibility of 
choice and turns action into consequence. This does not limit the 
author to any particular attitude, happy or unhappy, but it does assume 

1 84 The Denatured Novel 

that the individual has a hand in the vast circumstances of cause and 
effect. There is no such thing as a convincing deterministic novel. A 
novel may argue determinism, which is quite another matter, but an 
author overwhelmed by warfare can record only events without human 
cause, which is just what happens in the normal war novel. 

Authenticity cripples this stylized war fiction. It presupposes certain 
kinds of characters, it dwarfs them in their setting, and it dictates a set 
of attitudes conditioned by shock. These arbitrary conditions usually 
make the characters unbelievable. The soldier-author came to know 
the personnel and operations of some basic military unit: a squad, a 
bomber crew, or a watch-section aboard ship; but he also met many 
different kinds of men, far more than in any normal circumstances. 
He feels the obligation to make his story universal, and so one way or 
another he tries to confine his extensive, actual acquaintance to the 
fictional version of the military unit he knows. 

The actual citizen-soldiers have come from everywhere, representing 
every human quality, sharing only their amateur status. Some belonged 
to racial minorities, others were racists. Some grew splendidly, others 
remained weak and ineffectual. Some were bestial, some ignorant, and 
a few had quick, disciplined minds. They represented all regions and 
every social class. William March's narrator states his case and, un- 
wittingly, the problem. "This book started out to be a record of my 
own company, but I do not want it to be that, now. I want it to be a 
record of every company in every army/' 

How do you achieve a fiction which represents this actual range of 
human involvement in war? William March introduced 113 members 
of Company K, each with his own story to tell. His novel is a congre- 
gation of small voices, a succession of brief intimacies which leave only 
the impression of bewildering sameness. With the same war but with 
radically different proportions, John Dos Passos invested three fictional 
soldiers with heavy symbolic value; so symbolic that they rarely be- 
come individuals. 

Somewhere between these extremes the normal novel of World War 
II settles on a platoon cosmology. Some small, definable military unit 
contains a few permanent tenants and a great many transients. Four or 
five principal characters are carefully differentiated and then sur- 

Arms and the Manner 185 

rounded by a dozen or two dozen caricatures. The normal result is a 
crowded procession of strangers: the brave officer or the cowardly offi- 
cer, the shagbark sergeant with a deep sense of commitment, the Jew 
who becomes a better soldier than his Gentile persecutors, the bully 
and the bigot and the poet, the sensitive chaplain, and the cynical 

This is the cosmology in Leon Uris' novel, Battle Cry. The book is 
an encyclopedic record of a radio company in the second Division of 
the Sixth Marines: from civilian life, through boot camp and further 
training, to combat in the Pacific War and the climactic battle in the 
landings on Saipan, then back to civilian life. Despite the accumula- 
tion of day-to-day details the characters never develop or change. In a 
remarkable variety of circumstances each new disclosure merely man- 
ages to reveal more of the same. The reader knows these individuals no 
better than the author once knew their actual counterparts. 

These characters demonstrate the sentimental assumption that hu- 
man beings are good (with the author's solicitous attention most of 
them decide in favor of chastity or marriage). Beyond this, however, 
one can judge them merely as soldiers; excellent Marines, to be sure, 
but nothing more. When some of them die, terribly and violently, the 
fact has nothing to do with their individuality, but only with where 
they happen to be when the enemy shell explodes. Because the reader 
never knows them, he must respond to death or survival in merely 
quantitative terms. Life and death in this novel evoke about as much 
emotion as the lyric to 'Ten Little Indians." 

The same cosmology marks another normal novel: Kenneth Dod- 
son's Away All Boats ("The mightiest novel of men and war since 
Battle Cry"). This novel tells of a war vessel, an attack transport, in 
the Pacific theater. Through the small-boat officer and his crew the 
novel introduces the rest of the ship's personnel: on the bridge, in the 
ward room, by the landing net and on the beachhead. The confusion 
and stress and misdirected energy of a landing operation are effectively 
amplified by the dozens of walk-ons and extras engaged in haste and 
violence, a fact congenial to the filming of the story in Vistavision. 

The persons on this seascape are forever appearing about their busi- 
ness and disappearing again, but they have no consequence. The ship's 

i86 The Denatured Novel 

Captain Hawkes becomes eccentric as the pressure mounts. He keeps 
a pet monkey, for which he reserves all his intimate disclosures. He 
commands his ship-fitter to build a sail boat, and scavenges the fleet for 
red ink to dye its sails. His obsession for sunsets periodically paralyzes 
him, and when he conns his ship in such a trance he nearly sinks it. 
Hawkes is no character in his own right, but he is a pale reflection of 
Herman Wouk's Captain Queeg. As in The Caine Mutiny much is 
made of this eccentric man jeopardizing the entire command. Then 
suddenly the narrative denies the conflict it has elaborately built up; 
Hawkes sustains a brain concussion and dies in his bunk. 

Executive Officer Quigley is another Queeg-like character who turns 
out to be an anti-climax. He has made a career of evading responsibil- 
ity. His deftness at this threatens to involve some of the junior officers 
in a crisis with the captain, but in the events of the ship's last cam- 
paign somehow nothing comes of it. Then, there is an ensign aboard 
who schemes for favors at the expense of his colleagues, but after sev- 
eral dark hints about his malevolence, the author excuses him from the 
novel. These truncated characters all come to nothing. Events out- 
grow them, and they have no cause and no effect. 

The characters are casualties of the novel's limiting authenticity. 
This book only pretends to be about human beings, while it makes 
them merely components of a military unit. Its only necessity comes 
from the chronology of military events, the sequence of landing opera- 
tions from Makin to Okinawa; it stops only when the Navy has run 
out of beachheads. Like Battle Cry (which was temporarily "the 
mightiest novel of men at war") Dodson's book wages a war of attri- 
tion on its characters. 

The details of these books concenter events, not characters. This 
explains the pervasive but unlocalized setting of the normal war novel. 
All the training camps and ultimately all the beachheads are indistin- 
guishable because events are reconstructed in terms of that great com- 
mon denominator, military procedure: the officially prescribed method 
of carrying out any operation. Procedure constitutes the setting of 
these narratives regardless of their location. 

Just as procedure prescribes the action of a soldier's life and all the 
methods of his trade, so in these novels it guides the reader in the tech- 

Arms and the Manner 1 87 

nicalities of soldiering. Procedure is the basis of all military communi- 
cation, the means by which the individual remains a part of his unit 
and his unit a part of the entire organization. Military crises can best 
be understood in the terms of procedure, particularly procedure gone 
wrong, which is why it becomes such an insistent metaphor in these 

Sometimes this metaphor accomplishes wonders. The Good Shep- 
herd (1955), by C. S. Forester, tells the story of seventy-two hours in 
an Atlantic convoy, sometime in 1942; it dramatizes the convoy's con- 
tinuous battle with a pack of German submarines, seen from the bridge 
of a destroyer. Details of the battle are refracted through the Captain's 
mind: incessant trigonometry problems which jostle his recollections 
of Ecclesiastes and Job, and earlier moments in the Captain's own life. 
An interior monologue punctuates the narrative which advances by the 
sequence of orders on the bridge and by conversations between vessels 
on the TBS radio circuit. The counterpoint (of procedural informa- 
tion and one man's recollections) makes a convincing picture, but 
Forester's achievement was unusual. 

The enabling metaphor of procedure more often interferes with the 
characters, and burdens the story with irrelevance. The final episode 
in Dodson's Away All Boats, after the ship has been hit by kamikazes, 
necessarily becomes an urgent problem in damage control, but by then 
it is too late to salvage the story. Too many earlier episodes have merely 
demonstrated procedure for its own sake. Long before its climax the 
novel has become an instruction manual. 

Occasionally a novel, like Wirt Williams' The Enemy (1951; 1956), 
exists exclusively in terms of procedure. This tells of the cruise of a 
destroyer in a task group ordered to hunt Nazi submarines in the Atlan- 
tic prior to the invasion of Normandy. It enlarges on the irony of 
cruising 20,000 miles to achieve only an oil slick on the ocean the 
only evidence of the enemy. But the fiction appears to exist merely for 
the sake of demonstrating shipboard routine and the procedure of anti- 
submarine warfare. 

This demonstration specifies the details of a watch schedule, the 
prescribed methods of steaming in formation and communicating be- 
tween ships, relieving the watch and taking the conn, the mechanics 

1 88 The Denatured Novel 

of a search plan for an enemy submarine, the operation of the sonar 
stack, and the arming and launching of depth charges. Everyone 
mouths the same esoteric vocabulary of compass readings, commands 
to the wheel, code names in the security procedure, and all of the con- 
versational short-cuts between crew members. In his fascination with 
these magic terms the author even stops to define the accepted kenning 
for a bologna sandwich. 

The Enemy thoroughly demonstrates the mechanics of bearing and 
range without measuring anything human. Beyond their procedural 
vocabulary the characters have no existence, but this is not even an 
issue in the novel. First acquaintance establishes several facts: that the 
executive officer has a wife and prefers merely to drink while on liberty, 
that the captain is capable and has red hair, that the gunnery officer 
likes jazz and gunnery problems, and that the cohabitations of another 
officer become a public joke. This is all. Even the officer who hides in 
the depth-charge locker and sucks his thumb is merely a more insistent 
caricature than the others. 

The paperbound edition of The Enemy quotes Ernest Hemingway 
as saying, "A first-rate novel of the way it really was/ 7 That about sums 
it up. The urgency to explain "the way it really was" can overpower a 
fiction. It can dismiss cause and effect in favor of merely how it hap- 
pened. Presumably the urgency in these ponderous narratives is the 
imminence of death, but even here they contradict themselves. 

Death in the normal novel convinces more than life does. The de- 
tails of dying receive loving attention. The suppurating corpse and the 
still-warm body punctured by a single bullet are meant to suggest the 
range of horror, just as the typical characters are meant to suggest the 
range of commitment, but the novels achieve only shock and contra- 
diction. The soldiers with the most to live for die. They suffer death by 
irony; and by this insistent fact these novels imply some force which 
disqualifies moral being. 

Occasionally a war novel that is not quite normal will make a point 
of moral judgment. Fred Majdalany's Patrol (1953) demonstrates the 
awful discrepancy between the terrors of the isolated few who are actu- 
ally committed to battle and the many hangers-on behind the line. It 
tells of a hazardous and unnecessary reconnaisance, ordered in effect 

Arms and the Manner 1 89 

by the headquarters mess officer who was anxious to please a general. 
Among the patrol's casualties, the company's commander has fought 
on the lines so long and so continuously that his body simply wears out, 
unable to sustain a superficial leg wound. He dies by irony. 

Ned Calmer's The Strange Land (1950; 1951) insists on the same 
ugly comparison between the values of the combat soldier and the 
values of the spectator. This novel describes an Allied attack upon the 
Siegfried Line, launched under the worst possible conditions merely to 
gratify a major general's personal and professional advancement. The 
higher the rank and the farther away from the enemy, the more crim- 
inally irresponsible men become. 

A comparable but different crisis of morals occurs in Francis Irby 
Gwaltney's The Day the Century Ended (1955; Between Heaven and 
Hell, 1956). On the titled day, somewhere in the Philippine jungle, 
the hero realizes the inadequacy of the chivalric ideal which his Na- 
tional Guard commander has taught him. He has also learned to hate 
the brutish cruelty of the American army and the prison company to 
which he has been summarily transferred. The book makes no choice 
between confused ardor and organized hatred, but disqualifies them 
both in the hero's singularly moral concern over human behavior. 

Still another narrowly defined case against warfare is Walt Sheldon's 
Troubling of a Star (1953; 1954), about the American Air Force in 
Korea. By the actions of three of its characters, two pilots and their 
commanding officer, this novel evaluates the public concept of hero- 
ism and finds it fraudulent. The brave pilot commits an act of compas- 
sion for which he is grounded and branded a coward. The cowardly 
pilot hides his own ineptness by commiting suicide in a dive attack on 
an enemy tank by public standards an act of heroism. Their com- 
manding officer, who has sought preferment with napalm bombs and 
public relations, badly mistakes the courage of one pilot and the cow- 
ardice of the other. Having seen enemy soldiers burning up after one 
of his unit's missions, this colonel retires himself from service and 
accepts a job with the church after first punishing the compassionate 

The novels of Sheldon, Gwaltney, Calmer and Majdalany are all 
humanistic, having jettisoned the equipment of the normal war novel. 

190 The Denatured Novel 

They all happen to be negative in attitude, but that is beside the point. 
Gwaltney's is the best of them, although each one, in its narrow 
didactic version of human failure, has abandoned the chance for anal- 
ogy, for development. Ideally, the normal novel should be able to de- 
velop its platoon-organization into a convincing system of analogies, 
but this is a rare achievement. 

Anton Myrer accomplished it in The Big War (1957; 1958). This 
novel carries the usual full pack: members of a squad in a marine com- 
pany assigned to a landing operation on a Pacific island; episodes in the 
lives of three of these soldiers, from their terminal leaves in the United 
States, during their passage across the Pacific, through their beach- 
head landing to the final days of the battle on the island. Yet Myrer 
has achieved something different from the mere shock and denial of 
the normal war novel. The experiences of his three main characters 
dramatize a growing perception of moral man and the cost of morality. 

In The Big War the son of a Greek millhand, an ebullient Irishman, 
and a proper Bostonian reveal disparities in the presumed New Eng- 
land conscience. The moral man is Danny Kantaylis, the millhand's 
son. This Greek has gone over the hill to marry a girl who carries his 
child and the burden of her mother's neurotic hatred of him. He has 
already refused a veteran's privilege of retiring to an easy duty with 
war-bond rallies, and he returns to camp to suffer imprisonment and 
beatings for his absence-without-leave so that he can rejoin his squad 
to go overseas. 

Outward bound on a spaceless, impressionistic voyage across the 
Pacific, the men in this squad begin to emerge, refracted through the 
disdainful consciousness of the Bostonian, Alan Newcombe. Broken in 
rank Kantaylis will nevertheless lead the men; and his over-developed, 
uncompromising, and ultimately fatal sense of responsibility begins to 
dawn on Newcombe. The refraction of this through Newcombe's mind 
is the burden of the book. As this young patrician gropes toward some 
understanding of Kantaylis' excessive commitment, he begins a confes- 
sional to the girl he took to bed for one night and never saw again : 
letters written but not mailed, composed but not written a painful, 
ludicrous spectacle by which Newcombe grows a capacity to feel and 
to love. 

Arms and the Manner 191 

The narrative bears out Newcombe's struggle over the terms of the 
:omplicity between man and God. The battle repays the soldiers in 
:ind. With the squad cut off in the jungle, Kantaylis leads an impos- 
ible counter-offensive against an enemy tank, which he destroys even 
is it destroys him. Newcombe's final perception of this act of compas- 
;ion precedes his own violent death. Among the squad's survivors, 
ilong with the bestial mortals whose ignorance of compassion has in- 
;ured their safety, are several slow but educable men and Jay O'Neill, 
vho perceives instinctively what Newcombe had to struggle to learn. 

Back home, in The Big War, the next-of-kin offer the same bewil- 
lered motley of response to the fact which governs this fictional world: 
:hat the moral man's commitment kills him. Compassion saves the 
Awld, yet it cannot preserve itself against what it saves. As in Kantay- 
is' death and in the birth of his natural child, compassion begets life 
then dies in the birthing. Like any war novel, The Big War acknowl- 
edges the ironies of death and life, but it also assumes the complicity of 
the individual in the scheme of things. Its appraisal of the consequence 
Df moral man, in a world at war or not, very nearly explains the ironies 
it poses. 

Most normal novels deny the participation The Big War demands. 
They show sequence without consequence, violent death by merely 
external means. Their characters demonstrate humanitarianism and a 
capacity for misdemeanors, but none of the passion or the pride or the 
purpose that cause tragedy. Most of them conclude, long before the 
end, that human responsibility does not exist and that moral struggle 
is irrelevant in the face of the implacable force of war. And worse, they 
take inefficacy for granted. 

These all are signs of mannerism. The normal war novel evolves by 
the borrowing of literary conventions from previous novels. These con- 
ventions can make a book superficially similar to a predecessor enough 
for advertising purposes but they make no sense by themselves; the 
argument which would validate them has been left out. The enabling 
dialectic of these mannered novels exists not in themselves but in 
earlier books. Most contemporary soldier-authors have thoroughly ac- 
cepted their predecessors' tentative conclusions, and what were once 
thematic characteristics have become ornamental. 

1 90 The Denatured Novel 

They all happen to be negative in attitude, but that is beside the point. 
Gwaltney's is the best of them, although each one, in its narrow 
didactic version of human failure, has abandoned the chance for anal- 
ogy, for development. Ideally, the normal novel should be able to de- 
velop its platoon-organization into a convincing system of analogies, 
but this is a rare achievement. 

Anton Myrcr accomplished it in The Big War (1957; 1958). This 
novel carries the usual full pack: members of a squad in a marine com- 
pany assigned to a landing operation on a Pacific island; episodes in the 
lives of three of these soldiers, from their terminal leaves in the United 
States, during their passage across the Pacific, through their beach- 
head landing to the final days of the battle on the island, Yet Myrer 
has achieved something different from the mere shock and denial of 
the normal war novel. The experiences of his three main characters 
dramatize a growing perception of moral man and the cost of morality. 

In The Big War the son of a Greek millhand, an ebullient Irishman, 
and a proper Bostonian reveal disparities in the presumed New Eng- 
land conscience. The moral man is Danny Kantaylis, the millhand's 
son. This Greek has gone over the hill to marry a girl who carries his 
child and the burden of her mother's neurotic hatred of him. He has 
already refused a veteran's privilege of retiring to an easy duty with 
war-bond rallies, and he returns to camp to suffer imprisonment and 
beatings for his absence-without-leave so that he can rejoin his squad 
to go overseas. 

Outward bound on a spaceless, impressionistic voyage across the 
Pacific, the men in this squad begin to emerge, refracted through the 
disdainful consciousness of the Bostonian, Alan Newcombe. Broken in 
rank Kantaylis will nevertheless lead the men; and his over-developed, 
uncompromising, and ultimately fatal sense of responsibility begins to 
dawn on Newcombe. The refraction of this through Newcombe's mind 
is the burden of the book. As this young patrician gropes toward some 
understanding of Kantaylis 7 excessive commitment, he begins a confes- 
sional to the girl he took to bed for one night and never saw again : 
letters written but not mailed, composed but not written a painful, 
ludicrous spectacle by which Newcombe grows a capacity to feel and 
to love. 

Arms and the Manner 191 

The narrative bears out Newcombe's struggle over the terms of the 
complicity between man and God. The battle repays the soldiers in 
kind. With the squad cut off in the jungle, Kantaylis leads an impos- 
sible counter-offensive against an enemy tank, which he destroys even 
as it destroys him. Newcombe's final perception of this act of compas- 
sion precedes his own violent death. Among the squad's survivors, 
along with the bestial mortals whose ignorance of compassion has in- 
sured their safety, are several slow but educable men and Jay O'Neill, 
who perceives instinctively what Newcombe had to struggle to learn. 

Back home, in The Big War, the next-of-kin offer the same bewil- 
dered motley of response to the fact which governs this fictional world: 
that the moral man's commitment kills him. Compassion saves the 
world, yet it cannot preserve itself against what it saves. As in Kantay- 
lis' death and in the birth of his natural child, compassion begets life 
then dies in the birthing. Like any war novel, The Big War acknowl- 
edges the ironies of death and life, but it also assumes the complicity of 
the individual in the scheme of things. Its appraisal of the consequence 
of moral man, in a world at war or not, very nearly explains the ironies 
it poses. 

Most normal novels deny the participation The Big War demands. 
They show sequence without consequence, violent death by merely 
external means. Their characters demonstrate humanitarianism and a 
capacity for misdemeanors, but none of the passion or the pride or the 
purpose that cause tragedy. Most of them conclude, long before the 
end, that human responsibility does not exist and that moral struggle 
is irrelevant in the face of the implacable force of war. And worse, they 
take inefEcacy for granted. 

These all are signs of mannerism. The normal war novel evolves by 
the borrowing of literary conventions from previous novels. These con- 
ventions can make a book superficially similar to a predecessor enough 
for advertising purposes but they make no sense by themselves; the 
argument which would validate them has been left out. The enabling 
dialectic of these mannered novels exists not in themselves but in 
earlier books. Most contemporary soldier-authors have thoroughly ac- 
cepted their predecessors' tentative conclusions, and what were once 
thematic characteristics have become ornamental. 

192 The Denatured Novel 

Tracing back along this line of descent is not very difficult. Most 
normal war novels stylize the books which created the contemporary 
market: The Naked and the Dead, The Young Lions, and From Here 
to Eternity. Each of these earlier books has a dialectic. Each searches 
out the cause of inefficacy and argues determinism. These books, in 
turn, develop an argument which already existed in John Dos Passos' 
novel, Three Soldiers, which is the archetype of most contemporary 
war novels. 

Military organization grinds Dos Passos 7 three soldiers to extinction 
without regard for their defiance or their futile self-indulgence. All 
three have joined the AEFs crusade to France in the Great War. But 
the war ends halfway through the book, and the Army which preceded 
the war and which remains after it is the force which crushes them. 
One soldier remarks of their common plight, "It's part of the system. 
You've got to turn men into beasts. . . ," 

The force categorically denies their ambition to be persons. The sol- 
dier who merely wants to make corporal and to lead a squad against 
the enemy contracts a venereal infection; he is court-martialed, sen- 
tenced to a labor battalion, and assigned to permanent KP. Another 
soldier wants to distinguish between his obligation to the Army and 
his right to deny the sadism of its petty officials. But the system goads 
him into murdering his immediate adversary, and he spends the rest of 
his short life as a fugitive. The third soldier responds to the system's 
organized pieties ("which glorify personal greed and fear and hatred") 
by deserting ("to join the forlorn men, to throw himself into enviable 
defeat, to live life as he saw it in spite of everything"). He is hunted 
down and consigned to a brutal and permanent captivity. 

The novel specifies each individual's complicity in his own downfall. 
Each soldier in his way realizes his partnership with circumstance. But 
the action is refracted successively through the minds of these victims, 
so the most obvious impression is the blame of the system. The Army's 
arbitrary denial is so consistent, and chance is so prevalent, that the in- 
dividual's plight seems utterly insecure. 

Contradiction rules this particular world. Warfare itself is inconclu- 
sive, and death by combat is ironically incidental. The only deaths 
made much of in the book are a murder, a suicide, and a heart failure 

Arms and the Manner ' 193 

(suffered by a recruit who dies in bed, having refused to answer mus- 
ter). The military enemy is scarcely seen, but the omnipresent enemy 
is the Army. When a company clerk, a Jew named Eisenstein, says as 
much about the system's need to turn men into beasts he is court- 
martialed for his disloyalty. He has already been tried by the rest of 
his outfit and found guilty of being a Jew, a contributing factor to his 
later indictment. 

The normal war novels of the present generation have updated the 
irony of Three Soldiers in a peculiar way. They insist on a similar denial 
of the individual, but they ironically overlook the fact that each of Dos 
Passos' three soldiers complies in his own downfall while blaming the 
system. They faithfully exaggerate the details but ignore the reason in 
them. Thereby they contradict the whole contention of Three Soldiers 
that the individual is the agent of his own dereliction. 

This has happened by stages. The best sellers which established the 
contemporary market The Naked and the Dead and The Young 
Lions exaggerated certain features in Three Soldiers. All three novels 
document events which defeat the individual, but Mailer and Shaw 
make a deterministic theory of it; and both create a dialectic about the 
individual and the sovereignty of the system. Their dialectics are so 
ponderous, however, that later novels have merely exaggerated the 
symbols and the narrative details and left out the thesis. What has 
happened, therefore, is determinism by default. 

The Naked and the Dead and The Young Lions are the most doc- 
trinaire of all contemporary war novels. The determinism of The 
Naked and the Dead is primarily secular. This novel tells of the inva- 
sion of a Japanese-held island; its principal characters are a general, a 
lieutenant, and the members of an intelligence-and-reconnaisance 
platoon. For each character the author has thematically joined past and 
present by a device he calls "The Time Machine": ten impressionistic 
biographies (in the present tense) which punctuate the narrative (ren- 
dered in the past). The war merely intensifies whatever it is that has 
already limited each individual. From this unpleasant past each char- 
acter brings to the war a kind of undirected dislike which grows under 
new pressures into casual hatred. 

But the accumulated past is only a secondary cause of dehumaniza- 

194 The Denatured Novel 

tion. Something called "the medium" is the prime cause and the 
subject of the most speculation. The lieutenant sees his own analogy 
in the floating seaweed, the giant, rootless kelp which absorbs its 
nourishment from the medium around it. The general tries to find a 
metaphor of life and death in the trajectory of an artillery shell the 
asymmetrical parabola, rising gradually and abruptly falling. The "re- 
sistance of the medium" the atmospheric pressure causes the sud- 
denness of the fall. But the general tries to apply this metaphor to all of 
experience: the resistance of the medium causes an abrupt descent 
analogous to the "tragic curve." Whatever "the medium" is it cannot 
be controlled. 

At this point the unresolved speculation drifts back into the narra- 
tive, which "proves" the general's analogy. Nothing that happens to 
end the campaign or the lives of the characters has anything to do 
with human cause. A fatal patrol action turns out to have been unnec- 
essary in view of a victorious frontal attack, but the attack itself has 
succeeded because an ignorant staff officer has mistakenly deployed his 
troops. This particular military operation has merely intensified the 
denial of one's own efforts; it has merely sharpened all of the unfocused 
groupings without relieving or changing a thing. 

The Young Lions also insists on an unknowable determining force. 
The book is less oracular than Mailer's, but it obviously contradicts 
itself. Shaw invokes a vindictive statement from the Old Testament 
book of Nahum, in which God promises to devour the young lions, the 
war makers of the earth. But the execution of this promise in the novel 
denies the justice of the Old Testament God. 

The divine prophecy involves five young lions in particular: three 
Germans and two Americans. One idealistic German becomes a 
capable soldier and then a predatory animal; he is finally shot to death 
without ceremony. Another German, a professional killer, loses his 
face in an explosion and dies by suicide. The attrition of war itself 
apparently validates God's prophecy and accounts for these two deaths. 
But the story of the other young lions strangely qualifies the prophecy. 
One of these is an American Jew assigned to an infantry company of 
Southern Whites; he must fight even to live, let alone to find accept- 

Arms and the Manner 195 

ance. Then, having won his struggle at enormous cost, even as he cele- 
brates the compassion of another human being, he too is killed. 

This soldier is shot in the midst of his eulogy to "the human beings" 
who will conquer the animals in the world. This is the novel's final im- 
pression; it is no casual irony. Halfway through the book a German 
soldier has similarly been murdered in the midst of his plan for peace 
and his plea for humanity. "We have to show the world that there are 
still human beings in Germany, not only animals," he says just before a 
crazy series of events snuff him out. This illogic defies the God of 
Nahum. The technicality of being a young lion apparently prevails 
over the humanism of these two soldiers. 

The only principal character left standing at the novel's conclusion, 
in fact, has apparently escaped God's promise by his own ineptness. He 
is like the uncommitted, ineffectual majority of men. He has enlisted 
in the army to support a vague ideal and conveniently to escape his 
domestic entanglements. Failing to become an officer, he has arranged 
a transfer from hazardous duty, and his entire career is a record of aim- 
less irresponsibility. 

The scheme by which this man lives and the others die is impene- 
trable. If the author intended this soldier to represent the Old Testa- 
ment man divinely ordered to witness and believe, then he has bur- 
dened the God of Nahum with unbelievable subtlety. This inept sol- 
dier, by the structure of the novel, does not even know of the life and 
death of most of the others. Some system of value based on chance or 
implacable denial rules this fictional world. Like the unresolved mus- 
ings of the general Mailer's spokesman in The Naked and the 
Dead, the contradiction of The Young Lions focuses attention on 
some nameless determining force. 

James Jones's deterministic novel published three years later, con- 
tains some vestiges of a dialectic. The agents of faceless force are the 
generals who see "all men as masses, as numerical groups of Infantry, 
Artillery, and mortars that could be added and subtracted and under- 
stood on paper." So says the sergeant who contentedly regards himself 
as "the instrument of a laughing Providence." Another soldier, a pri- 
vate named Prewitt who has denied the Army's total claims upon him, 

196 The Denatured Novel 

tries to formulate the reason for his persecution: ". . . it was the system 
that was at fault, blame the sysem. But he could not even blame the 
system, because the system was not anything, it was only a kind of 
accumulation, of everybody, and you could not blame everybody, not 
unless you wanted the blame to become diluted into a meaningless 
term, a just nothing." 

As in Dos Passos' novel, this hard-headed private of James Jones has 
complied in the events which bring him down and finally kill him; and 
so have all the other characters: a soldier who gives up and resigns from 
the system by accepting the charge of insanity; the sergeant who ex- 
ploits the system for his own ends; the company commander who tries 
to make a prudent show of conformity and fails. But the novel aban- 
dons this humanistic premise, and buries it under a massive accumula- 
tion of events which overtake the characters. The inefficacy of these 
soldiers is just as much the subject of Jones's novel as it is of Mailer's 
or of Shaw's, although Jones's Private Prewitt is more sophisticated 
than the others, more informed about what will happen to him and 
more willfully compliant, less enraged or bewildered. This reflects the 
fact that Jones, unlike the other citizen-soldiers, belonged to the regu- 
lar Army. 

The economic importance of these three contemporary books antici- 
pated the similarities of the normal novels that followed them. Battle 
Cry, The Enemy, Away All Boats, Able Company, and dozens like 
them all packed the same equipment without the reason for it: the 
reportage, the heavy documentation of setting, the maze of procedure 
which confines large numbers of small human beings as though de- 
tails were intrinsically worthwhile. These stylized novels report the 
facts of warfare without the heat of thought. They have no antago- 
nism; they fail to search out the design of things because they have 
abandoned any notion that the complicity between events and the in- 
dividual might work both ways. 

The war novel has calcified about as much as it can in Lawrence 
Kahn's Able One Four. This book abridges the normal novels so effi- 
ciently that it reads like an index. According to its paper cover it cele- 
brates "an unheralded and heroic branch" of the service, as almost all 
of them do. Its characters are the crew of an armored vehicle engaged 

Arms and the Manner 197 

in the Allied advance across Germany. After a glimpse at the crew 
members, the author briefs the reader on the procedure of waging war 
in a tank destroyer, then gets right to the climactic battle. It is all 
very orderly. 

Procedure rules this handbook and confines what passes for charac- 
ters. Procedure is the insistent setting of the tank destroyer itself: "Ele- 
vation six-four-eight," "left one-two-seven," "fire for effect," "up five 
hundred"; "Super wants you at oboe time. Repeat, oboe time. Over."; 
watchword and countersign, H.E. and A.P. shells; "TOT starts at zero 
five four five"; K-rations and "five-in-ones," and "bedcheck Charlie"; 
"Wilco, out." 

There is not much else to say about the five enlisted men and their 
lieutenant. The college graduate suffers from battle fatigue, has bad 
dreams, and obsessively writes letters to his wife. A more well-adjusted 
member of the crew has no wife and did not go to college. As for the 
others, the lieutenant is overworked, the coward is unpleasant, and 
the "tight-lipped" sergeant benignly rules the crew. The last man is a 
replacement, an "innocent kid" who diligently learns his new job as 
gun loader. When the tank destroyer finally gets shelled, there is a 
characteristic irony about the destinies of its crew: the college man 
goes into shock, the coward gets off with a flesh wound, and the inno- 
cent youngster violently dies; his corpse on the road beside the tangled 
vehicle the book's final impression literally fulfills the epigraph on 
its paper cover: "Blood . . . guts . . . and hot steel!" 

Able One Four is legion on publishers' lists. This abridgement of the 
normal novel is about ciphers, not people; it is an extremity of literary 
naturalism, but not the only one. This tyrannous fashion even sub- 
jugates novels which start out to be humanistic. The Good Shepherd, 
by C. S. Forester, is typical of the contradiction. This is the book which 
makes procedure so convincing, but it ends up with some unassimi- 
lated second thoughts which deny the narrative itself. 

The problem of getting the convoy through, in The Good Shepherd, 
lies with the escort commander; the narrative makes this clear. His 
skill and training and sense of responsibility oppose a formidable yet 
equally human enemy. Although the issue is in doubt, the terms of the 
battle are finite, even professional. This scarcely fits the attitude which 

198 The Denatured Novel 

has sold so many war novels, but Forester's conclusion makes amends. 
At the end of the convoy the captain throws himself on his bunk while 
his exhausted mind sums it all up: "Chance the chance that elevated 
a paranoiac to supreme power in Germany and a military clique to 
power in Japan dictated that when it was too late he should receive 
the coveted promotion to commander, if it can be called chance." No 
ending could deny its narrative more arbitrarily than this one does. 
"Chance had made him an orphan; chance had brought about the 
senator's nomination. Chance had put him in command of the convoy 
escort. Chance had made him the man he was and had given that man 
the duty he had to carry out." 

Stylizing has caused the scarcity of humanistic war novels even un- 
convincing ones. Putting characters into situations for the sake of situ- 
ations deprives the characters of cause and effect; it forfeits their 
responsibility in what happens to them. The more a given novel is 
concerned with human complexity, therefore, the more its caricature 
is likely to deny it. 

The most influential of the war novels posing an ethical problem has 
been Herman Wouk's The Caine Mutiny. It so happens that Wouk's 
novel also disqualifies the ethical problem he poses. He built a shrewd 
combination of acceptable attitudes, more adroitly than C. S. Forester 
later did, but The Caine Mutiny is at least based on the human capa- 
bility of causing or avoiding a particular crisis. Because of its tremen- 
dous commercial success it was later caricatured just as strenuously as 
were the novels of Mailer and Shaw and Jones. And the stylizations of 
Wouk's ethical novel were just as deterministic as the normal war 
novels turned out to be. 

Despite all the books unreasonably likened to The Caine Mutiny 
one obvious choice has gone begging. This is Martin Dibner's novel, 
The Deep Six (1953; 1954)- The same publisher issued both books a 
year apart. Dibner's book is an obvious case of total recall, and the 
publisher must have felt he could not serve the sales of either novel by 
advertising this fact. Superficially, Dibner's book is more like The 
Caine Mutiny than The Caine Mutiny is. It exaggerates every pre- 
dominant characteristic of its predecessor and therefore arrives at a 
totally different state: it is a gross caricature of mechanical men. 

The Caine Mutiny offers a case history of Commander Queeg, a 

Arms and the Manner 199 

ship's captain burdened by bad luck, incompetence, and the symptoms 
of mental illness. His officers build their grievances into a convincing 
case for relieving him of his command. After the executive officer has 
accomplished this legal mutiny he is acquitted in a court of law 
through the shrewd tactics of his defense attorney. Then the novel 
suddenly turns on itself. 

The defense attorney indicts the ship's officers for their moral tres- 
pass, but for reasons irrelevant to the trial or to the narrative. The 
defense attorney is a Jew with grievances against the Nazis, who boiled 
his mother into soap. He ignores Queeg's pathology for his own. He 
applauds all the Queegs, all of the professional soldiers who stood 
guard against the Nazis while these reserve officers were leading a fat, 
happy life. 

Wouk's spokesman indicts the mutineers on the ground that they 
are accountable for their irresponsibility a condition from which he 
pardons Oueeg at the same time. Either his pardon or his indictment 
is illogical. Actually, the author has shuffled his attitudes so that nearly 
everybody wins. All of the major characters get an endorsement and a 
reprimand. The novel's only major casualty is the enlisted man whom 
Queeg has driven insane. 

Certain situations in the narrative of this ethical problem have un- 
dergone an obvious sea change in The Deep Six. They have become 
bloated. Instead of an ancient destroyer, the ship in the latter book is 
an ancient cruiser; it floats more incompetence, more sadism, and more 
tyranny than the destroyer. A curious thing happens to the demonic 
qualities. The cowardice, the vindictive cruelty, the mental illness and 
moral trespasses implied of Queeg are multiplied in the latter novel; 
they become explicit in the persons of four different officers in Dibner's 

The captain of this cruiser has no concept of his own command, and 
like Oueeg he is suspected of incompetence by his superiors. The gun- 
nery officer, like Quccg, is an outright coward. The executive officer is 
a psychopathic hater; like Queeg he issues degrading commands for the 
sake of personal reprisal, and like Queeg he is relieved of his duties. The 
malevolence which Queeg's character implies becomes a fact in the 
character of an ensign aboard the latter ship. 

Violence is always threatened aboard the Caine, but it explodes 

200 The Denatured Novel 

aboard the cruiser. Racism is only a belated topic in Wouk's novel, 
merely asserted after the mutiny and left undeveloped, but the perse- 
cution of a Negro become a major factor in Dibner's book. Homosexu- 
ality is insinuated in The Caine Mutiny, in Queeg's persecution of his 
helmsman; Wouk used it to establish the complexity of Queeg's para- 
noia. But in The Deep Six homosexuality is a blatant fact; an officer's 
trespass upon an emotionally disturbed enlisted man embellishes that 
novel's climax. 

The exaggerations multiply wherever you look. Wouk tells a separate 
story in the love affair of the junior officer who survives the court-mar- 
tial to become the Game's last captain. To satisfy his mother he has 
forsaken a girl whom he later decides he loves. This story measures the 
man's growing up. 

The love story in The Caine Mutiny is a way of giving the character 
credentials, for it is his reflections on the mutiny which are allowed to 
stand. But the love story in The Deep Six, also involving a girl from 
across the tracks, exists merely for itself. Her body is simply one of the 
many temptations in the way of virtue. The junior officer is also a 
painter who is tempted by the financial gain of drawing cheap illustra- 
tions, and he is a Quaker deeply involved in the hatred aboard ship. 
But he has already made up his mind before the novel starts: he stands 
resolutely for painting, marriage and good will. It is ever so easy and 

All of these exaggerations lose point in the second novel. The Caine 
Mutiny is concerned with the nature of its central character, and it 
presents a crisis of moral responsibility. But The Deep Six merely 
demonstrates a number of moral felonies. To be sure, Wouk resolved 
a crisis to evferyone's advantage, but Dibner merely shocks; he offers 
nothing to resolve. The gross irony of The Deep Six is that it has noth- 
ing to fathom. 

Dibner's stylization of The Caine Mutiny is on a par with Robert 
Bowen's stylization of From Here to Eternity, in his novel called 
Bamboo. Both Bowen's and Dibner's novels negate cause and effect; 
they offer separate sensations but no sensational analogies of a conflict. 
They negate the very properties by which a novel can appraise experi- 
ence. There is only an academic difference between these books and 

Arms and the Manner 201 

Lawrence Kahn's Able One Four. Preceded by so many normal war 
novels, Kahn's book employs conventions which were cliches to begin 
with. But there is no point in comparing degrees of unbelievability. 

Most normal novels exploit shock and denial, but do not develop it 
to any avail. These novels offer sequence without consequence. For 
that illusion of necessity which all fiction must have they depend 
merely on details of actual experience and on the conventions of other 
novels. Because they fail to make convincing characters they disqualify 
the whole system of moral choice without which, why fiction? 

All of these authors of normal war novels offer passive, abhuman 
characters the better if passive, because if they act on their own they 
are not human enough to make it believable. This particular species of 
fiction shows no evolution of form. "Devolution" would suit it better. 
The normal war novels, for the most part, breed abhuman characters in 
an abhuman world. 

Chapter IX 

Just Deserts 

THE BOOK TRADE denatures novels in an effort to sell more of them, and 
publishers excuse themselves on the grounds that they cannot give the 
people "better" than the people want. Publishers simplify their wares, 
invest them with palatable attitudes, and make them easily digestible; 
then they advertise them as vital and life-giving. But they cannot have 
it both ways; a novel's vitality lies in its being complex enough to ap- 
proximate life. 

Most contemporary business novels and war novels commit this 
fraudulent simplism. They are not the only offenders, but they are 
representative. The normal war novel burdens the characters with 
consequences they did not cause and cannot change. Because of the 
incapacity of these characters nothing much is at stake. The system 
prevails. With the metaphor of business instead of warfare, denatured 
novels resolve the conflict more pleasantly but just as irrevocably in 
favor of the system. The businessman is confined by the great corpora- 
tion which employs him, but it is all for the best. The benevolent sys- 
tem will take care of him. 

Under the book trade's auspices these novels add witless authority to 
such determinism. They create characters for ready-made situations 
dictated by the success of previous books like the narratives, the char- 
acters have been denatured. But there is even more to this denial of 
moral struggle. The aim to gratify quickly makes a novel serve an atti- 
tude. Thinned down to a narrative which must be resolved, the de- 
natured novel is susceptible to any attitude. It becomes contrived, yet 
this is what commercialism has forced upon writers. 


Just Deserts 203 

The peculiar task of the denatured novel is to furnish some expedient 
reciprocity for all the excesses it contains in the first place. It is "moral," 
as the publishers say, because every character gets his just deserts. This 
reciprocity masquerades as justice, which it is not; but like most in- 
genious devices it will satisfy anyone who takes it for granted. The 
device approximates an eye for an eye, but it exploits an astigmatism 
in this old and tested retribution. What happens is more nearly an eye 
for a tooth, which is something else again. 

Expedient reciprocity, for example, is what makes The Caine Mu- 
tiny come out even. Almost every character gets what is coming to him 
even a little more for good measure, which is immediately satisfying. 
The trick is to balance a character's debits and credits from all sources 
and to pay him appropriately in a lump sum. This seems so obviously 
tidy, and it avoids tedious delays of the sort actual people have to en- 

Wouk's novel excuses Queeg for his trespasses on the grounds that 
he is sick; it even applauds him for being a sick man who has stayed on 
duty (although it only implies that the sickness was service-induced). 
The book also indicts Lieutenant Keefer who has interpreted Queeg' s 
irresponsibility to the rest of the officers and managed the charges of 
mental illness against the captain. Keefer might be sick too (he is cer- 
tainly obsessed with Queeg), but Keefer is held responsible as an indi- 
vidual and Queeg is pardoned. 

The book obscures this moral lapse by making punishment and re- 
ward seem reciprocal. During the trial the defense attorney, Green- 
wald, has just destroyed whatever esteem Queeg had, so in a fit of re- 
morse he publicly does the same for Queeg's tormentor. He insults 
Keefer by throwing a drink in his face and blaming him for the affair. 
This insult does not justify Queeg any more than Queeg's defeat has 
justified Keefer, but it gratifies the wish that everyone should get his 
just deserts. 

In order to make this reciprocity stick (and it does stick; Keefer is 
overcome by a sense of defeat), the author must validate Greenwald 
in some way. He invents a past for Greenwald (which has nothing to 
do with Keefer) to justify Green wald's insulting somebody or other. 
His Jewish mother in Germany was killed by the Nazis, so he trades 

Chapter IX 

Just Deserts 

THE BOOK TRADE denatures novels in an effort to sell more of them, and 
publishers excuse themselves on the grounds that they cannot give the 
people "better*' than the people want. Publishers simplify their wares, 
invest them with palatable attitudes, and make them easily digestible; 
then they advertise them as vital and life-giving. But they cannot have 
it both ways; a novel's vitality lies in its being complex enough to ap- 
proximate life. 

Most contemporary business novels and war novels commit this 
fraudulent simplism. They are not the only offenders, but they are 
representative. The normal war novel burdens the characters with 
consequences they did not cause and cannot change. Because of the 
incapacity of these characters nothing much is at stake. The system 
prevails. With the metaphor of business instead of warfare, denatured 
novels resolve the conflict more pleasantly but just as irrevocably in 
favor of the system. The businessman is confined by the great corpora- 
tion which employs him, but it is all for the best. The benevolent sys- 
tem will take care of him. 

Under the book trade's auspices these novels add witless authority to 
such determinism. They create characters for ready-made situations 
dictated by the success of previous books like the narratives, the char- 
acters have been denatured. But there is even more to this denial of 
moral struggle. The aim to gratify quickly makes a novel serve an atti- 
tude. Thinned down to a narrative which must be resolved, the de- 
natured novel is susceptible to any attitude. It becomes contrived, yet 
this is what commercialism has forced upon writers. 


Just Deserts 203 

The peculiar task of the denatured novel is to furnish some expedient 
reciprocity for all the excesses it contains in the first place. It is "moral," 
as the publishers say, because every character gets his just deserts. This 
reciprocity masquerades as justice, which it is not; but like most in- 
genious devices it will satisfy anyone who takes it for granted. The 
device approximates an eye for an eye, but it exploits an astigmatism 
in this old and tested retribution. What happens is more nearly an eye 
for a tooth, which is something else again. 

Expedient reciprocity, for example, is what makes The Caine Mu- 
tiny come out even. Almost every character gets what is coming to him 
even a little more for good measure, which is immediately satisfying. 
The trick is to balance a character's debits and credits from all sources 
and to pay him appropriately in a lump sum. This seems so obviously 
tidy, and it avoids tedious delays of the sort actual people have to en- 

Wouk's novel excuses Queeg for his trespasses on the grounds that 
he is sick; it even applauds him for being a sick man who has stayed on 
duty (although it only implies that the sickness was service-induced). 
The book also indicts Lieutenant Keefer who has interpreted Queeg's 
irresponsibility to the rest of the officers and managed the charges of 
mental illness against the captain. Keefer might be sick too (he is cer- 
tainly obsessed with Queeg), but Keefer is held responsible as an indi- 
vidual and Queeg is pardoned. 

The book obscures this moral lapse by making punishment and re- 
ward seem reciprocal. During the trial the defense attorney, Green- 
wald, has just destroyed whatever esteem Queeg had, so in a fit of re- 
morse he publicly does the same for Queeg's tormentor. He insults 
Keefer by throwing a drink in his face and blaming him for the affair. 
This insult does not justify Queeg any more than Queeg's defeat has 
justified Keefer, but it gratifies the wish that everyone should get his 
just deserts. 

In order to make this reciprocity stick (and it does stick; Keefer is 
overcome by a S2nse of defeat), the author must validate Greenwald 
in some way. He invents a past for Greenwald (which has nothing to 
do with Keefer) to justify Greenwald's insulting somebody or other. 
His Jewish mother in Germany was killed by the Nazis, so he trades 

204 The Denatured Novel 

on this deserved compassion by throwing a drink at the tormentor of 
one of the men who opposed the Nazis. This all happens so quickly 
that it hides the illogic, and the reader settles accounts at a tooth for 
an eye. 

This sort of pay-off never works in actuality, which is why it is so 
satisfying in fiction. When a character accrues a large balance of credit 
he can squander it all, if the author likes, in some violent attack upon 
some other character; the more violent the pay-off the more satisfying. 
In this way an author can excuse one excess by another and still main- 
tain the pretense of being moral. But excess is the real subject. 

Virtually every war novel trades on its subject; The Caine Mutiny 
is conservative in this respect. War itself licenses dishonesty, violence, 
and trespass. The novelist can declare himself impeccable simply by 
claiming that this is the way it was; he is merely calling a spade a spade. 
Peacetime mores are often irrelevant in wartime, so when he heeds the 
call of the spade he can dig deeper in the dirt and get away with it. Or 
so he rationalizes. 

War does cauterize the routine experiences into which it cuts; no 
doubt it is to be judged differently. But this does not relieve the novel- 
ist of the need to make his situation believable. If anything it limits the 
bounds of probability, but most war novelists do not see it this way. 
Walt Sheldon's novel, Troubling of a Star, for example, obviously ex- 
ceeds its license. This is a novel about the cowardly pilot and the brave 
pilot and their commanding officer in Korea. It is pointedly didactic, 
but this does not hide its indulgence. Of the five other characters in 
the novel, two of the women arc nymphomaniacs and the third is pleas- 
ingly amoral. One of the men ( the only enlisted man ) is a homosexual, 
and the other is a black marketeer who keeps one of these imperious 

There, is plenty of violent reciprocity among these trespassers. The 
homosexual is ground up and spit out of an airplane propeller; one 
nymphomaniac murders her keeper and the other speeds the cowardly 
pilot to his flaming death. The expedient violence evens out the debits 
and credits, which clearly exist for the sake of the violence. This does 
not even begin to convince, because the novel's selectivity is so obvi- 
ously contrived. The world is simply not populated in this way. 

Just Deserts 205 

Excess is not limited to felony and perversion. Even chastity can be 
too much. In Francis Gwaltney's war novel, The Day the Century 
Ended, the soldier-hero indulges in a long recollection of his honey- 
moon, of the consummation of his marriage, and of subsequent achieve- 
ments of this union. This soldier's preoccupation is understandable, 
but it does not happen to concern the action; it is gratuitous which is 
one thing love ought not to be. Chaste or unchaste, any situation which 
exists for its own sake is an extravagance no fiction can afford; it de- 
stroys the whole proportion. From the total impression of any novel, 
nothing secedes like excess. 

The war novels have no monopoly on excess. That ultimate business 
novel, The Empire and Martin Brill, by George DeMare, presents 
the same sort of crapulence. This is the book in which the anonymous 
corporation rules all of its employees. The blurb on its back paper cover 
asserts there is another story in this novel, in the "heart and knowledge 
of Martin Brill/' "fascinating, sardonic and perverse." This is an ac- 
count of Brill's sexual experience with the "lovely April who could give 
him her body but withhold her heart." 

This marketese refers to the hero's frustrations over a woman who 
would lie with him even as she reminded him of her charity. But the 
"lovely April" has no other function. The flesh larded into this narra- 
tive has no cause and shows no necessity. The "lovely April" is a paper- 
bound heroine imposed on the story, and one reason why the book is so 
scattered. She is unbelievable, which is her only connection with the 
rest of the novel. 

In the flood of business novels which deluged the market after Mar- 
quand's Point of No Return, Cameron Hawley's Cash McCall gave a 
special notoriety to the conflicts in buying and selling companies. This 
subject in turn begat more derivative books, including some which 
were not business novels at all but merely disguised to look like them. 
John MacDonald's paperback original, Man of Affairs (1957), for ex- 
ample, is incidentally about a raid on a company for the sake of manip- 
ulating its stocks, jut the book's real subject is a violent house party in 
the Bahamas at which the scoundrel tries to secure some signed proxies. 

The extravagance of Man of Affairs includes bribery, blackmail, 
fornication and violent death, all tricked up to seem as wholesome as a 

206 The Denatured Novel 

Girl Scout cookie. The excess is rationalized by expedient reciprocity. 
The narrative keeps tabs on what every character has coming to him, so 
that each excess apparently justifies another. For example, one un- 
pleasant character gets ripped apart by a barracuda; and worse, he is 
eaten up as a bonus for being beaten up by the hero. 

It takes some India-rubber logic to justify this sadism, particularly 
since his only misdemeanors were insulting his wife and then the 
hero. But the extravagant pay-off is arranged in easy stages. This un- 
pleasant character wanted to embarrass the hero after that stalwart had 
beaten him up so he went skin diving with the barracuda; he had been 
beaten up because he had picked the fight; and he had picked the fight 
in the first place to justify his having insulted the hero (this was his 
real mistake committing two misdemeanors in a row, without letting 
the hero have his turn ) . 

This travesty on retribution typifies the paperback violence on sale 
in every cigar store. It is a peculiar derivative of journalism and slick 
magazine fiction sold in the same store. Journalism is a point of view 
as well as a technique of writing. It is a kind of selectivity. The reporter 
chooses a timely subject; more important, he stresses whatever is un- 
usual about it he finds an angle. It is the exception to the routine 
which makes the news story, and millions of readers are used to it. 

The exceptional slant is also a legitimate fictional device, particu- 
larly if the intended readers also read newspapers. But the journalistic 
point of view in fiction has its hazards; fiction does not have the au- 
thenticity of fact it cannot take consent for granted. Instead, it must 
provide its own justification; the invention must convince. The more a 
fiction exploits the unusual, therefore, the more patently it must justify 
it. The more excessive the action, the more difficult this feat becomes. 

The feat in journalistic fiction is to present an action with its ration- 
alization already in the narrative but not immediately apparent. In 
fact, it is characteristic of such fiction that it jumps off feat first. This, 
of course, is the magazine technique. The slick writer's trick is to pre- 
sent some apparently unsolvable conflict and then dissolve it through 
the sudden emphasis on some latent factor which has been present all 
along. He merely corrects a misunderstanding he has allowed to exist. 
The happy ending of most magazine stories is built in to begin with. 

Just Deserts 207 

Whether the extravagance is a happy ending for a magazine or a vio- 
lent sensation for a paperback, the sleight-of-hand is about the same, 
and the greater the shock the greater the satisfaction. There are many 
euphemisms for the technique of making an action acceptable. Maga- 
zines refer to "editorial values," playwrights call them laws of dramatic 
action, and professors of literature refer to conventions, but they are 
still the tricks of the trade. The trick in writing paperback fiction is to 
excite the reader with excess in such a way as to placate or elude his 
judgment which would otherwise condemn it. 

Paperback novels are aimed at particular kinds of retail outlets: 
drugstores, newsstands and cigar-and-candy stores. With these outlets 
it is natural that novels particularly original paperbacks should ex- 
ploit the techniques of the newspapers and the magazines with which 
they compete for attention. What is interesting is that technique in all 
of these media is so much like the gimmick in the movies: the sudden 
reversal which pays off every character in kind, The appeal of this fic- 
tion is the moral posture which excuses itself without stooping to do so. 

It is no coincidence that so many of James M. Cain's novels became 
movies and appeared in paperback. Cain just happens to represent this 
fiction whose declarative subject does not much matter and whose real 
subject is violence. The eclectic war novel or business novel or med- 
ical novel or hillbilly novel or whatever are merely more extreme ver- 
sions. These denatured novels merely detour their narratives into one 
sensation after another, with an expedient sham and the pretense of 
being moral. 

Sham and expedience arc considerably older than the novel, but 
their authority has been increased in this particular form of writing by 
two recent developments in American literary history. One develop- 
ment, of course, concerns the new and various auspices of subsidiary 
markets. The magazines, the films, and the paperbacks have all in- 
creased the efficiency of rationalizing the reader's indulgence. But an- 
other development, which is literary, has made this new efficiency pos- 
sible. In fact, the change in the novel form since the end of the nine- 
teenth century has become utterly congenial to the denaturing process. 
It has made the novel more easily exploitable. 

This major change in the form of the English novel has occurred 

208 The Denatured Novel 

within the past hundred years, accelerating in the American novel dur- 
ing the past thirty. It is the author's withdrawal from his own fiction: 
abdicating from the position of the moral observer and leaving such 
commentary to the characters. The novel has gradually changed from a 
dramatically illustrated essay to a dramatic presentation of a limited 
point of view. 

The crucial stage of this large development is commonly credited to 
Henry James, who learned to build a story around a character's attempt 
to define his predicament. For James the process of the character's dis- 
covery constituted the entire story. No sooner had James developed 
the convention of refracting incidents through the mind of a single 
character, however, than other novelists began to popularize it. There 
were different ways to appropriate the limited point of view, but the 
great popularizing movement was actually a retreat from James's sub- 
ject of the mind itself. 

Reading James's novels is hard work; but the fiction of Edith Whar- 
ton, Anne Sedgwick and Ellen Glasgow simplified the involvements 
and the emotions by describing them. These writers had learned a les- 
son from the master, and more people were able to read (and buy) 
what they wrote. This particular stylization, however, of substituting 
description for the dramatic illusion, has continued to the point of 
utter boredom. Herman Wouk's Marjorie Morningstar (1955) is a 
fair sample of the extremity. 

But other American writers developed the dramatic point of view in 
other ways. James had argued, in his famous essay on "The Art of Fic- 
tion/' that art is a personal, direct impression of life. Impressionism 
was already a fact when James wrote this, and the development of 
impressions instead of commentary was another way of removing the 
author from the scene. After awhile the overseer of the fictional world 
was no longer even a character. This is the narrative fashion which so 
much popular fiction has caricatured during the last thirty years, 
clearly under the influence of Ernest Hemingway. 

Hemingway was a style-maker long before he won the Nobel Prize in 
1954, before the Swedish Academy cited him "for his powerful style- 
forming mastery of the art of modern narration." He has said that 
"style" is the way you got something done; it is the simplest and there- 

Just Deserts 209 

fore the best way of doing it. His fiction demonstrates it. Hemingway's 
stories are apparently irreducible. The narratives are simple and ex- 
plicit, but so understated that their signficance is always a matter of 

The characters rarely testify about the encounter which makes a 
story, and the events themselves stop short of resolving it. Heming- 
way's short stories usually end just at the point where the conflict be- 
comes explicit, or where the character realizes the conflict. The im- 
portance of this moment in the life of a character had been the partic- 
ular interest of Sherwood Anderson, from whom Hemingway absorbed 
a great deal. 

Anderson's interest was introspective; he attempted to show some 
past crisis in the character's life which would account for his present 
nature. He built his story according to what he called the "history of 
moments." This critical "moment" is also the subject of most of Hem- 
ingway's short stories, but with a difference. For Hemingway the story 
is the process of making this moment finally explicit. 

Indirection and understatement mark Hemingway's fiction; they 
emphasize the irony which is usually his subject. These devices are also 
what the stylizations of Hemingway exploit and exaggerate, and the 
result is usually empty and pretentious. The cohering principle of 
Hemingway's stories is the realization of the way things are. The end is 
implicit in the beginning; it merely and finally becomes explicit. There 
is no other resolution. But caricature can destroy this principle. 

Hemingway explained his purpose and its hazards in his autobio- 
graphical essay, Death in the Afternoon (1932). This primer on bull- 
fighting is also an essay on style. Hemingway began by saying that he 
"wanted to put down what really happened in action; what the actual 
things were which produced the emotion that you experienced." He 
wanted to show the "how" of things so exactly that the "why" would 
be apparent. "The real thing," he said, was "the sequence of motion 
and fact which made the emotion," and which with luck would en- 
dure. In this "sequence of motion and fact" his stories cohere. 

The sequence omits emotion, By stressing the physical sensations 
which accompany emotion, however, Hemingway renders the moment 
of crisis by impressions of inconsequential things. When Frederic 

2 1 o The Denatured Novel 

Henry leaves the hospital, in A Fare-well to Arms, his child dead and 
his lover dying, his mind records the precise facts about a plate of ham 
and eggs. In The Sun Also Rises, when Lady Brett discusses her gen- 
erosity toward another man with Jake, the sexual cripple she would 
rather have had, Jake orders a martini. The man dying of blood poison- 
ing in "The Snows of Kilimanjaro" tries to appraise the fatal turn of his 
misspent life, and he recalls the details of the picture he was trying to 
take with his camera when he fatally scratched himself. 

By this process the dying man finally gets the picture of his life, 
which is what Hemingway's fiction is all about. A few characters get 
the picture with distorted clarity; and others never get it at all. Nothing 
is resolved, but the clarity of "the actual things . . . which produced the 
emotion that you experienced" gradually emerges. It emerges by in- 
congruity: by the omission of what you might expect to find. 

In Death in the Afternoon Hemingway wrote "The dignity of the 
movement of an iceberg is due to only one-eighth of it being above 
water." To define the submerged mass he plots what he can see. He is 
like the painter who sees the "negative" space around the object it con- 
figures. As the chasm dramatizes the cliff, the silence completes the 
experience. "If a writer of prose knows enough about what he is writ- 
ing ... he may omit things that he knows," Hemingway declares; "and 
the reader, if the writer is writing truly enough, will have a feeling of 
those things as strongly as though the writer had stated them." 

But this is where stylization fails to convince. Hemingway himself 
anticipated it: "A writer who omits things because he does not know 
them only makes hollow places in his writing." Whether there are 
"hollow places" or pregnant ones, in this style of indirection, depends 
on whether sensations exist for their own sake or to serve the narrative. 
Exaggerating the ironic understatement or incongruous omission de- 
natures the novel form, because it leaves no overseer in the fictional 
world: no presence to preside, no scheme or pattern to explain itself. 

Sensation for its own sake can destroy the indirection of fiction. In 
the extreme it even denies the complex nature of human experience, 
as the narratives of Paul Bowles illustrate. Bowles describes predica- 
ments which arise out of nowhere and lead violently to nothing. The 
Sheltering Sky (1949) is about a white woman in the Sahara desert 

Just Deserts 211 

who is raped to the point of insanity. Let It Come Down (1952) is 
about an American who hammers a nail into the ear of his Arab vic- 
tim for no apparent reason. That is all there is to these narratives. 

"The Distant Episode," a short story in The Delicate Prey (1950), 
weighs just as much as any one of Bowles's novels and fairly summa- 
rizes all of his fiction. In this story a nameless professor is attacked by 
Arabs who cut out his tongue, throw him into a saddle bag, and keep 
him as a pet; they dress him in tin plates and teach him obscene 
dances. After a year or so they sell him, and at this point the professor 
recovers sufficiently from his stupor to break away and run out into the 
desert by himself. There is nothing else. 

This is the end of the line. Experience becomes merely inexplicable 
sensation. Paul Bowles's mindlessness is not worth considering, except 
that it shows the absolute vacuum which an author's exit from his fic- 
tion can accomplish. Into this sort of vacuum, in the absence of any 
cohering scheme, some slick variation of tit for tat can be made to pass 
for moral judgment. This is where journalism becomes relevant to the 
novel, along with commercially sponsored sham and expedience. 

Publishers add the pretense of morality to fiction that is already 
mindless. When they piously say they must give the people what they 
want, they mean that people want sensation for its own sake, along 
with some face-saving device for indulging in it. As both an editor and 
an advertiser, therefore, the publisher who makes the pious disclaimer 
usually encourages the point of view which justifies one sensation by 

The reprinter's advertising of Erskine Caldwell's mindless fiction is 
one example of this; so is the advertising of James M. Cain's. The dif- 
ference between these two writers is that Cain himself exploits the 
moral pretense; he has misused" the sort of narrative indirection which 
Hemingway developed. But even Cain is modest compared to the lum- 
bering distortions of John O'Hara. 

O'Hara is a self-conscious legatee of Hemingway's "style-forming 
mastery of the art of modern narration." He has eulogized Hemingway 
as "the outstanding author out of the millions of writers who have lived 
since 1616"; and he has acknowledged his own indebtedness to Hem- 
ingway. His short stories, like Hemingway's, advance to an explicit dis- 

212 The Denatured Novel 

closure of the conflict and then stop. But they are more obvious and 
more insistently ironic than Hemingway's, and more limited in sub- 
ject. They usually clarify the inept rationalization of a weak, unpleas- 
ant character; and they manage a precise twist at the end, in the man- 
ner of the short short story. O'Hara excels in this form. He is so adept, 
in fact, that he conforms all his novels to it. 

John O'Hara writes the longest short short stories in the literature. 
His first and best book, Appointment in Samana, is usually praised for 
its brevity and efficiency. The book dramatizes the events which pre- 
cede a suicide. It is a terrifying story, no doubt of it; but mostly be- 
cause it never explains itself. It lacks that novelistic quality of apprais- 
ing as it represents its subject. 

Appointment in Samana chronicles events during the last three 
days in the life of a suburban businessman: his social trespasses which 
embarrass or insult his peers, and which presumably cause him to kill 
himself. Each trespass is credible, but the author's problem is to make 
them seem like a chain of events. In this he fails, for they do not reveal 
why the hero did what he did. Instead, the book offers the expedient 
reciprocity: the churl gets his just deserts, and the ending says, "I told 
you so." 

A short story and a novel satisfy differently. A single incident can 
end convincingly in one of many ways; but when a series of incidents 
substantially repeat one another, as in a novel, each is bound to help 
explain the rest, and the number of convincing resolutions decreases. 
The simple shrug of irony will not do for a novel. This hero kills him- 
self after his self-esteem has suddenly caved in. His trespasses are symp- 
toms of this fact, but why or how they are connected with his appalling 
inability to stay alive the author ignores. 

O'Hara even obscures the question with a series of flashbacks which 
have no relevance to events in the present. For example, the hero's 
grandfather embezzled a lot of money and then blew his head all over 
the hayloft; but neither embezzlement nor suicide is hereditary. As 
a boy the hero had once stolen a flashlight from a store ,and been 
caught; but this scarcely justifies his suicide over money matters. The 
hero's faithful wife once had some unpleasant sexular adventures be- 
fore she met the hero, but they are obviously irrelvant to his problem. 

Just Deserts 213 

None of these alarums and excursions show cause; but they do inter- 
rupt the narrative and confuse the sequence of "fatal" events enough 
to obscure its illogic. For this reason the story succeeds as well as it 

O'Hara keeps telling the same story that E. A. Robinson superbly 
told, of "Richard Cory" who put a bullet through his head for no 
known reason. O'Hara adds nothing but words. His most celebrated 
novel, Ten North Frederick (1955), tells of another suicide, with two 
important changes: it is longer and more pretentious than his first 
novel. It tells about a well-born, capable citizen who drinks himself to 
death. All the circumstances of this character's erosion are explicit, yet 
the novel invalidates all of the lengthy reasons that might have 
caused it. 

The hero's wife despises him after realizing she cannot ever own 
him; but the hero is scarcely aware of this. He loses two million dollars 
in the stock market; but he has one million dollars left. He sustains a 
broken leg and a concussion; but they heal in plenty of time for him to 
begin drinking himself to death. Much is made of his frustrated polit- 
ical ambitions; but he shows every resolve to regain his self-confidence, 
and says as much. 

The book minimizes the only uncontradicted reason why this hero 
might prefer cirrhosis of the liver. He despatches his daughter's mar- 
riage by annulment and abortion. He does so in the space of three 
sentences: about seventy words of the novel's 140,000. Despite the 
novel's exhaustive testimony about his love for his daughter and his 
sense of responsibility in general, there is not a word about the soul- 
agony this experience might have caused him. In one sentence, fifty 
pages after the event, the hero tells his lover (his daughter's room- 
mate) that he thinks he might not have done the right thing. This is 
the sort of "hollow place" Hemingway warned about. In O'Hara's case 
it is a vast emptiness which completely begs the character. 

O'Hara's other major efforts are equally pretentious. A Rage to Live 
(1949) amplifies the sexual instinct in its heroine. The first half of this 
inordinately long book sounds like a mating manual, as it trains the 
heroine for marriage, and the second half demonstrates the uncon- 
vincing foolishness of this assertedly shrewd woman. In tedious detail 

214 The Denatured Novel 

it describes her just deserts and how she earns them, but why she earns 
them is no concern of O'Hara's. His latest book, From the Terrace 
(1958), is even more fulsome and more evasive. It is a complete sta- 
tistical record of irrelevant details in the life of an investment banker 
and his community. This pompous exhibition even includes footnotes, 
yet it completely shuns the responsibility of making a believable fic- 

O'Hara's endless narratives are monuments to moral pretense. They 
fake the novel's complexity with the simplism of total recall. All fic- 
tion distorts, and a novel's genuine complexity can make distortion 
into a commentary on more than itself. But a novel's only justification 
is its relevance to actuality; if it is not believable in this sense, it is 
worthless. The trouble with most denatured novels like O'Hara's is 
that they are unbelievable, so whatever pretense they make is gross. 

The ideal of a mindless pretense to morality is what the marketese 
of the book trade tries to create. Because fiction has always been con- 
sidered a luxury in this nation, it has nearly always been sold in a 
buyer's market and disguised as a useful instrument. The economic 
history of the trade since the 1 890*5, and particularly since the 1930'$, 
has merely aggravated the difference between what seems and what is. 

The distinguishing fact about most of our national fiction is that it 
is published for non-readers. This nation now has more literate citizens 
who do not read than any nation in tne world. All publishers Know 
this, which is why most of them are afraid to take a chance. The best 
way to stay solvent is to satisfy the customer. But the customer is rarely 
a reader; he is merely a consumer. 

In every retail industry the consumer is king. He is the "average 
man" who is sought after and gratified. This is relevant to the eco- 
nomic organization of the book trade. The major changes in the pub- 
lishing industry since the end of the nineteenth century have all been 
changes in marketing methods, and they reflect the demise of the 
entrepreneur. The publisher is no longer in control. Much of his edi- 
torial function has been appropriated by the so-called "subsidiary" 
distributors of fiction via magazines, films, book clubs and cheap re- 
prints; and these distributors are committed to the mythical "average 

Just Deserts 215 

Book publishing always has been inefficient; the trade publisher has 
never really been in touch with the readers who buy novels. The book 
industry knows no such thing as that ideal of total communication 
which market researchers lovingly idealize; in their jargon the pub- 
lisher has no efficient "feedback" from book buyers. This is so because 
his real customers are not book buyers but marketers. 

When the bookstores return his "best sellers 7 ' by the carload, the 
publisher looks around for some signals. He gets them from the adver- 
tising of the book club which has turned down his last twenty books; 
from the reprinter who in turn responds to the magazine wholesaler; 
from the movie studio which takes its orders from film distributors and 
theater owners. In short, the trade publisher gets his signals from mer- 
chandisers and from the owners of entertainment properties who use 
fiction as a by-product. 

At this point the old antithesis between art and entertainment dis- 
solves into a synthesis called manufacture. From his marketers, the 
publisher derives an image of the tired housewife or the traveling busi- 
nessman or the stenographer waiting for her home permanent to set: 
consumers, who will pay to be gratified by fiction. Then, if the artifact 
happens to please, the publisher will search for others which reproduce 
it without seeming to. He manufactures reproductions. He also ration- 
alizes in public that he cannot give people any "better" than they 
want. But this is specious. The publisher does not know about the con- 
sumers of the subsidiary products any more than he might know about 
his readers. 

Although the publisher has become a frightened man, he still has a 
few vestiges of editorial authority. At least he has not yet reached the 
absurd predicament of the automobile manufacturer who lets his pub- 
lic relations staff design his products. The publisher has not yet choked 
on his own "feedback" because the trade is so badly organized (blessed 
are the disorganized). It has never been so centralized as the automo- 
bile industry or any other industry in the national economy. The dif- 
fusion of influence in the book trade has obstructed communications 
and delayed the tyranny of the consumer. 

In the present confusion the trade publisher still has a chance. At 
least he is the only one who can do anything about it. Now, as ever, 

2 1 6 The Denatured Novel 

his individual decision of whether to publish a given book or how to 
publish it, many times repeated, defines the quality of the literature. 
Literature may be no concern of the publisher, but it is his business. It 
so happens that the individuality of a book serves both literature and 
business. If the publisher can convince his marketers that the peculiar 
difference of every book is its most salable quality, there might yet be a 
revolution in the book trade which would give the reader his just 


Able Company, 180-181, 196 

Able One Four, 180, 196-197, 201 

Across the River and into the Trees, 142 

Act of Love, An, 179 

Adamson, Hans C, 138 

Alice Adams, 103-104 

All Quiet on the Western Front, 180 

All the King's Men, 146 

Ambassadors, The, 40 

American, The, 40 

American Magazine, 83-84 

American Mercury, 28 

American Publishers Association, 15 

American Spectator Yearbook, 109 

American Tragedy, An, 123-125 

Anderson, Sherwood, 85, 209 

Anthony Adverse, 159 

Ape s, Angels, and Victorians; The Story 
of Darwin, Huxley and Evolution, 158 

Apley, George, 173 

Appleton, D., Co., 27 

Appointment in Samarra, 145, 212 

Architects of Fate, 26 

"Are Novels Worth Reading?" 27-28 

"Are the American Novelists Deteriorat- 
ing?" 27 

Armat, Thomas, iio-m 

Arminian Heresy, 170 

Armory Show of 1913, 45 

Armstrong, Henry, 158 

Arrowsmith, 96 

Arrowsmith, Martin, 98 

"Art of Fiction, The," 208 

Arthur, Timothy Shay, 19 

Arthur Mervyn, 19 

"Artificial Selection," 84 

Asch, Scholem, 43 

Ashbaugh, George, 158 

Asylum, or Alonzo and Melissa, an Amer- 
ican Tale Founded on Fact, The, 17 

Atherton, Gertrude, 85, 108 

Atkinson, Donald T., 158 

Atlantic, 161 

Away All Boats, 181, 185, 187, 196 

Babbitt, 96-97 
Babbitt, Irving, 144 
Bacheiler, Irving, 108 
Baggott, King, 112 

Bailey, Temple, 85, 88, 100 

Baldwin, Faith, 85-86 

Balla, 45 

Bamboo, 181-182, 200 

Banning, Margaret Culkin, 86, 88, 89 

Bantam Books, 133, 140, 162 

Bara, Theda, 114 

Barlowe, Shep, 152 

Barrymore, John, 112 

Barwick, Cleves, 167-168, 174-175 

Barzun, Jacques, 73 

Battle Cry, 181, 185-186, 196 

"Bazaar Spirit," 27 

Beach, Rex, 108 

Beachhead in Hell, 

see Red Beach Assault 
Beer, Thomas, 89 
Behn, Aphra, 29 
Bellamy, Edward, 18 
Benet, Stephen Vincent, 91, 94 
Ben-Hur, 107-108 
Bennett, Arnold, 85 
Berle, Adolphe A., 166 
"Best Selling Consensus," 16 
Better Homes and Gardens, 84 
Between Heaven and Hell, see Day the 

Century Ended, The 
Bibbs, 1 01 
Bible, 25 

Big Fisherman, The, 43, 47 
Big War, The, 190-191 
Birth of a Nation, 121 
Bishop, John Peale, 92 
Blakesley, Roger, 162-163 
Blondes, Brunettes, and Bullets, 158 
Bluestone, George, 117 
Bolton, Mrs. Sarah Knowles, 26 
Book Find Club, 178 
Book-of-the-Month Club, 77-78, 91, 140, 

162, 178-179 
"Book Publishing: The Insular Industry, 

Book Supplement of the New York 

Times, 15 
"Book, the Buyer and the Critic, The," 

7i. 157 

Bookman, 15, 31 
Borzoi Quarterly, The, 159 
Bowen, Robert, 181, 200 

2 1 8 The Denatured Novel 

Bowker Lectures, 20, 48, 86, 89 

Bowles, Paul, 210-211 

Brandt, Carl, 91 

Breaking into the Movies, 119 

"Break-Up of the Novel, The," 27 

Brett, Lady, 210 

Bridget's Food of Life, 52 

Brief Retrospect of the Eighteenth Cen- 
tury, 13 

Brill, Martin, 175 

Bronte, Charlotte, 23 

Brown, Charles Brockden, 17, 19, 30 

Brown, Harry, 180 

Brush, Katherine, 88 

Buck, Pearl, 85 

Bullard, Avery, 170 

Bulwer-Lytton, Edward George Earl, ist 
Baron Lytton, 16, 23 

Burnham, James, 166 

Burns, John Home, 180 

Burton, Tony, 162-163 

Butterfly, The, 127, 130-131, 157 

Cabell, James Branch, 88 

Cain, James M., 126-132, 157, 207, 211 

Came Mutiny, The, 179, 181, 186, 198- 

200, 203-204 
Caldwell, Erskine, 136-137, 143, 148- 

154, 211 

Caldwell, Taylor, 85 
Call It Experience, 150 
Calmer, Ned, 189 
Calvinism, 24-25 
Carnegie, Dale, 26 
Carter, Ross, 181 
Cary, Lucian, 89 

CashMcCall, 167-170, 172, 174, 176, 205 
Castle, Everett Rhodes, 89 
Catalogue of All the Books Printed in the 

United States, 14 
Gather, Willa, 85, 88 
Censored: the Private Life of the Movies, 


Ceram, C W., 157 
Chaplin, Charlie, 112 
Charlotte Temple; A Tale of Truth, 17 
Chaucer, Geoffrey, 151 
Cheney, O. H., 49, 71-12, 79, 105, 144, 


Chicago Board of Trade Building, 176 
Chillingworth, 36-37 
Churchill, Winston, 178 
Cinderella, 34-36, 42, 91, 102 

Country Home, 83 

Country Lawyer, 63 

Covenant Theology, 25 

Covered Wagon, The, 88 

Crack-Up, The, 91 

Cram, Mildred, 109 

Crane, Stephen, 88, 122 

Crater, The, 18 

Crisis of the Film, The, 114 

Crockett, Chism, 153 

Crofts, Frederick, 20, 23 

Crowell, T. Irving, 25 

Crowell, Thomas Young, 25-26 

Crowell Corporation, 83 

Crusade in Europe, 178 

Cubism, 45 

Cummins, Maria, 23 

Current Literature, 27 

Curtis Publishing Company, 83-84, 140 

Dale, Edgar, 114 

Darwin, Charles, 122 

David Harum, 63 

Davis, Elmer, 86, 89, 104 

Davis, Richard Harding, 108 

Day the Century Ended, The, 189, 205 

Death in the Afternoon, 209-210 

Cinderella, 107, 120 

Civil War, 160 

Clark, Gil, 168-169, 174 

Clark, Walter Van Tiiburg, 145 

Cleavage Press, 58, 65 

Coates, Robert M., 147 

Cobb, Irvin S., 88 

Cohen, Octavus Roy, 89 

Coleman, Lonnie, 181 

Colliers, 83, 85, 96 

Company K, 183 

Complete Home Treasury of Successful 

Planning for Health and Happiness in 

Your Life, The, 52-53 
Compulsion, 142 

Conquest of Canaan, The, 101-102 
Conrad, Joseph, 85, 181 
Content of Motion Pictures, The, 114 
Cooper, James Fenimore, 16, 18, 21, 30, 47 
Copyright Act, 13-14, 16, 82, 136 
Coquette; A Novel Founded on Fact. 

The, 17 

Cosmopolitan, 84 
Costain, Thomas, 43 
Country Doctor, 63 
Country Gentleman, 83 



Deep Six, The, 198-200 

Defoe, Daniel, 29 

De Givry, Grillot, 158 

DeGraff, Robert, 135 

Deland, Margaret, 85 

Delicate Prey, The, 211 

Dell, publishers, 133, 140 

DeMare, George, 167, 174, 205 

"Devil and Daniel Webster, The," 91 

DeVoto, Bernard, 85, 89 

Dibner, Martin, 198-200 

Dickens, Charles, 16, 23 

Dimmesdale, The Reverend, 36-38 

"Distant Episode, The," 211 

Dodson, Kenneth, 181, 185-187 

Dodsworth, 40, 42-43, 47, 96-97 

Dodsworth, Sam, 41, 97 

"Does the Present-Day Fiction Make for 

Immortality?" 27 
"Dog on a Leash," 45 
Dollar Book Club, 77, 179 
Dooley, Mr., 83 

Dos Passos, John, 184, 192-193, 196 
"Double Indemnity," 129-130 
Douglas, Lloyd, 43 
Down-Easters, The, 20 
Dreiser, Theodore, 30, 88, 122-124, 134, 

1 60 

Duchamp, Marcel, 45 
Dunne, Finley Peter, 83 
Durable Fire, The, 168 

Ecclesiastes, Book of, 187 

Economic Control of the Motion Picture 

Industry, in 
Economic Survey of the Book Industry, 

49, 71-72, 79, 105, 157 
"Economics of Authorship, The," 86 
Edgeworth, Maria, 21 
Edison, Thomas, no 
Edmonds, Walter D., 89 
Eggleston, Edward, 19 
Eisenhower, Dwight, 178 
Eisenstein, 193 
"Embezzlers," 129 
Emerson, John, 119 
Eminent Authors Pictures Incorporated, 

Empire and Martin Brill, The, 167, 

174-177, 205 

End of the World, The, 19 
Enemy, The, 181, 187-188, 196 
Episode in Palmetto, 149 

Ernst, Morris, 116 

Ewald, William B., 158 

Executive Suite t 167, 170-172, 174, 177 

Fabliaux, 151 

Famous American Authors, 26 

Famous American Statesmen, 26 

Famous Men of Science, 26 

Famous Players Company, 113 

Farewell to Arms, A, 62, 109, 143, 180- 

181, 210 

Farm and Fireside, 83 
Farrell, James T., 30 
"Fate of the Novel, The," 27 
Faulkner, William, 88-89, I2 5, *34, l8 
Fawcett Publications, 140 
Ferber, Edna, 85, 88 
Fiction Book Club, 179 
Fielding, Henry, 29, 151 
First National Survey of Best Selling 

Books, 15 

Fisher, Dorothy Canfield, 85 
Fiske, Minnie Maddern, 112 
Fitzgerald, F. Scott, 30, 91-92, 94 
Ford, Corey, 89 
Forester, C. S., 187, 197, 198 
Foster, Hannah, 17 
Fletcher, John Gould, 114 
Franklin, Benjamin, 16, 25, 170 
Franklin Book Club, 58 
Franklin Library of Practical Living, 51- 

52, 65 

Frederic, Harold, 31 
Free Air, 96 
Free French Forces, 125 
French Underground, 125 
Freud, Sigmund, 122 
From Here to Eternity, 122, 179, 181- 

182, 196, 198, 200 
From Main Street to Stockholm, 95 
From the Terrace, 214 
Futurism, 45 

Galatea, 129 

Gale, Zona, 88 

Gallery, The, 62, 180 

Galsworthy, John, 85 

Gardner, Erie Stanley, 134 

Garland, Hamlin, 30 

Gaskell, Mrs. Elizabeth (Cleghorn 

Stevenson), 23 
Gathering Storm, The, 178 
Gaynor, Lewis, 48-50, 53-57, 60, 62-63, 

67-68, 70-80 

izo The Denatured Novel 

Gentleman from Indiana, The, 101 

Georgia Boy, 143 

"Girl Before a Mirror," 45 

Glamour, 84 

Glasgow, Ellen, 85, 88, 208 

Glory, God, and Gold, 158 

Gloves, Glory, and God, 158 

God, Gold, and Government, 138 

God, Sex and Youth, 158 

Gods, Graves, and Scholars, 157-158 

Gods, Heroes and Men of Ancient 

Greece, 158 

God's Little Acre, 136-137, 143, 149, 154 
Gods, Sex, and Saints, 158 
Goldberg, Rube, 44 
Golden Multitudes, 22 
Gold Medal Books, 133, 140 
Goldwyn, Samuel, 108 
Gone with the Wind, 32, 37-40, 47, 54, 


Good Housekeeping, 84 
Good Shepherd, The, 187, 197 
"Grains of Sands," 88 
Granlund, Nils T., 158 
Gray, Charles, 63, 148, 162-164, 166, 

168, 170, 172-173 
Great Gat shy, The, 92 
Greatest Trust in the World, The, 160 
Greenwald, 203 
Grey, Zane, 108 
Griffith, D. W., 107, 121 
Griselda, 154 

Grosset & Dunlap, 140, 162 
"Guardsman, The," 109 
Guest of Quesnay, The, 103 
Gulf Coast Stories, 149 
Gurko, Leo, 158 
Gwaltney, Francis Irby, 189-190, 205 

Hackett, James K., 113 

Hale, Edward Everett, 24 

Harcourt, Alfred, 95-96, 98 

Harcourt, Brace, 95 

Harland, Marion, 23 

Harlow, Jean, 114 

Harper's Bazaar, 84 

Harper's Magazine, 28, 95, 101 

Harper's New Monthly Magazine, 82 

Hart, James, 23 

Hawkes, Captain, 186 

Hawley, Cameron, 167-170, 172, 174, 

176-177, 205 
Hawthorne, Nathaniel, 18, 23, 30, 37-39 

Hayes Office, 116 

Hearst's International, 84 

Hemingway, Ernest, 32, 123, 125, 142- 

143, 180-181, 188, 208-213 
Henry, Frederic, 209-210 
Hergesheimer, Joseph, 89 
Heroes, Highbrows and the Popular 

Mind, 158 

Heroes, Kings, and Men, 158 
Herrick, Robert, 160 
History of American Magazines, 82 
Hollands, J. D., 180-181, 196 
Hollywood, 63, 106, 109, 112, 117, 132, 

177, 179-180 

Hollywood: The Dream Factory, 116 
Holmes, Mary J., 23 
Home-and-Jesus Formula, 22 
Hough, Emerson, 88 
House and Garden, 84 
House Beautiful, 84 
House of the Seven Gables, 18 
Howells, William Dean, 30, 81-82, 164- 


Huckleberry Finn, 143 
Huettig, Mae D., in 
Hughes, Rupert, 108 
Hulme, William E., 158 
Hurst, Fannie, 88 
Hynd, Alan, 158 

Ide, Hatcher, 101 
/ Wanted to Write, 92 
// Christ Came to Chicago, 24 
// Jesus Came to Boston, 24 
Image of Josephine, 103 
In His Steps, 24-25, 51 
Independent, 27 
Irvine, William, 158 
"Is Present-Day Fiction Quite Ephe- 
meral?" 27 

"Is the Novel Being Superseded?" 27 
"Is the Novel Done For?" 28 

Jack-and-the-Beanstalk, 101 

Jackpot, 150-151 

Jackson, Helen Hunt, 108 

Jacob, 165 

Jake, 210 

James, G.P.R., 21-22, 102 

James, Henry, 19, 40, 62, 90, 208 

Jane Eyre-Rochester Relation, 23 

Jessica, 148 

Jesus, 24-25, 43-44 

Index 221 

Job, Book of, 187 

John Brown's Body, 91 

Jones, James, 122, 179, 181-182, 192, 

195-196, 198, 200 
Journeyman, 151 
Judas Iscariot, 43 
Jungle, The, 160 

Kahn, Lawrence, 180, 196-197, 201 

KaJem Studio, 107 

Kantaylis, Danny, 190 

Kate Fennigate, 102 

Kearney, Patrick, 123 

Keefer, Lieutenant, 203 

Kennedy, John Pendleton, 21 

Kennicott, Carol, 97 

Kerr, Sophie, 89 

Kershner, Howard E., 158 

Key West, 124 

King, Basil, 108 

Knopf, Alfred, 157, 159, 162 

Korean Conflict, 180 

Koster and Bial's Music Hall, no 

Kyne, Peter B., 88 

Ladies' Home Journal, 27, 83-84, 162 
Laemmle, Carl, 112 
La Farge, Oliver, 89 
Lamplighter, 23 
Lapham, Silas, 164-166 
Lardner, Ring, 88 
Lasky, Jesse, 86 
Last Supper, 44 
Late George Apley, The, 92-93 
Lawrence, Florence, 112 
Lea, Fanny Heaslip, 88 
Leather Stocking Tales, 30, 107 
Legion of Decency, 115 
Lester, Jeeter, 143 
Let It Come Down, 211 
Levin, Meyer, 142 

Lewis, Sinclair, 31, 40-42, 85, 95-98 
Leypoldt, Frederick, 73 
Life of an American Fireman, 120 
Lippincott's Magazine, 27 
Literary Guild, 77, 100, 704, 179 
Little, Brown, 26-27 
Lively Lady, The, 92 

Lives of Girls Who Became Famous, 26 
Living Age, 27 
Lockwood, Charles A., 158 
Loesser, Thomas, 53, 55-56, 58-59, 61- 
62, 64, 67-70, 74 

Logan, A Family History, 19 

Lombard, Carole, 114 

London, Jack, 88, 90-91, 108, 122, 132 

Look Homeward, Angel, 62 

Looking Backward, 18 

Loos, Anita, 119, 125 

Lorentz, Pare, 116 

Lorenzo Bunch, The, 100 

Lorimer, George Horace, 87-89, 96, 98 

Louden, Joe, 101 

Love's Lovely Counterfeit, 130 

Ludwig, Emil, 85 

Macaulay, Richard, 89 

MacDonald, John, 205 

Macy, R. H., 15 

Madame Bo vary, 40 

Mademoiselle, 84 

Madison Avenue, 177 

Magazine Revolution, 82 

Magic, Myth, and Medicine, 158 

Mailer, Norman, 178-180, 194-196, 198 

Main Street, 95-98 

Majdalany, Fred, 188-189 

Managerial Revolution, The, 166 

Man from Main Street, 95 

Man in the Gray Flannel Suit, 167, 171 

Man of Affairs, 205 

"Man of Letters as a Man of Business," Si 

Mantrap, 96-97 

Mar hie Faun, The, 18 

March, William, 180, 183-184 

Marden, Dr. Orison, 26 

Marek, Kurt W., 157 

Marjorie Morningstar. 40, 208 

Marquand, John P., 30, 89, 92-93, 134, 

147-148, 156, 159, 161-168, 171-174, 

176, 182, 205 
Martin Eden, 90-91, 94 
Matthews, Brander, 53 
Maxwell-Essex Fund, 80 
McCall's, 84-85, 100 
McClure's Magazine, 160 
Mclntosh, Marion, 23 
Means, Gardiner C, 166 
Melies, Georges, 107, 120 
Melville, Herman, 30, 134 
Metalious, Grace, 106 
Midlander, The, 101 
Miller, The Reverend Samuel, 13-14 
Mirthful Haven, 103 
Miss Lonely hearts, 147 
Mister Roberts, 181 


The Denatured Novel 

Mitchell, Isaac, 17 

Mitchell, Margaret, 37 

Modern Corporation and Private Pro- 
perty, The, 1 66 

Moll Flanders, 40 

Money Writes! 84, 86 

Monsieur Beaucaire, 101 

Monster Books, 58, 64, 68 

Morgan, Harry, 124-125 

Morgan, J. P., 160 

Morina, 129 

Morris, Gouverneur, 108 

Moth, The, 131 

Motion Picture Academy Awards, 118 

Motion Picture Patents Company, no- 
112, 115 

Motion Picture Production Code, 116 

Motion Picture Research Council, 113-114 

Mott, Frank Luther, 22, 82 

Moving Picture World, 132 

Munsey, Frank, 94 

Murder, Mayhem and Mystery, 158 

"My Old Man," 143 

Myrer, Anton, 190 

Nahum, Book of, 194-195 

Naked and the Dead, The, 58, 62, 178- 

181, 192-196, 198 
Nation, 28 

National Book Award, 50 
National Usury Association, 63 
Nazarene, The, 43-44 
Neal, John, 16, 19-20, 30, 143 
Newcombe, Alan, 190-191 
New England Novels, 93 
New Leader, The, 73 
Newman, Christopher, 40 
New Testament, 43 
New York Herald Tribune, 63 
New York Times, 15, 63 
Night Before Dying, The, see Wisteria 


Nobel Prize, 42, 50, 95, 208 
Norris, Frank, 30, 88, 122, 134, 160, 176 
Norris, Kathleen, 85 
North American Review, 21, 27 
North of Grand Central, 93 
Notions of the Americans, 16 
Novels into Film, 117 
"Nude Descending a Staircase," 45 

O'Hara, John, 142, 145, 211-214 
O'Hara, Scarlett, 37-38, 42 

Peoples' Book Club, 179 

Perkins, Professor Horatio, 150 

Perkins, Maxwell, 61-62 

Peter Rabbit, 181 

Peyton Place, 106 

Phillips, David Graham, 88 

Picasso, Pablo, 45 

Pickford, 112, 114 

Pictorial Review, 84-85 

Pierre, 30 

Pike, Judge Martin, 101 

Pit, The, 176 

Plain Man's Pathway to Health, The, 52 

Plutocrat, The, 40, 42-43, 47 

Pocket Books, 133, 135 

Poe, Edgar Allan, 17, 21, 143 

Point of No Return, 63, 147, 161-164, 

166-170, 172-173, 176, 205 
Polk, Horace, 52-54, 70 
Polk, Horace, Jr., 68 
Polk & Franklin, 48-80 
Pollyanna, 108, 114 
"Poor Boys Who Became Famous," 26 
Poor Richard's Almanac, 25, 52 
Popular Book, The, 23 
Popular Library, 133, 140 
Porter, Edwin S., 120-121 
Porter, Eleanor, 108 
Porter, Gene Stratton, 31 
Olcott, Sidney, 107 
Old Testament, 131, 194 
Oliphant, Dan, 101 
Omnibook, 179 

One Hundred Years of Publishing, 26 
O'Neill, Eugene, 33, 181 
O'Neill, Jay, 191 
Oppenheimer, E. Phillips, 89 
Origin of the Species, 181 
Orphism, 45 
Ox-Bow Incident, 145-146 

Pamela; or, Virtue Rewarded, 16, 25, 40, 


Pandora, 127, 130 
Paramount Studios, 86, HI, 123 
Parsifal, 101 

"Passing of the Novel, The," 27 
Past All Dishonor, 127-128 
Patrol, 1 88 
Paul, Louis, 158 
Paulding, James K., 18, 21 
Peale, The Reverend Norman Vincent, 26 
Pease, Irvie, 101 

Index 223 

Post Hall Syndicate, 179 

Postman Always Rings Twice, The, 126- 

127, 129 

Powdermaker, Hortense, 116-117 
Power and the Prize, The, 167, 172, 175- 


Presenting Lily Mars, 103 
Prewitt, Private, 195-196 
Prynne, Hester, 35-39, 42 
Publishers' Weekly, 15-16, 27, 60, 63, 

70, 73, 79, i6, 178 
Pulham, H. M., 173 
Pulitzer Prize, 50 
Pushing to the Front, or Success Under 

Difficulties, 26 
Putnam, George Palmer, 23 

Queeg, Captain, 186, 198-200, 203 
Quigley, Executive Officer, 186 
Quo Vadis? 27 

Radcliffe, Anne (Ward), 21 

Radio Corporation of America, in 

Radio-Keith-Orpheum, in 

Rage to Live, A, 213 

Ramona, 108 

Randolph >e, 16 

Rath, Tom, 171-172 

Reader's Digest, 137 

Reader's Digest Condensed Book Club, 

78, 179 

Readers Guide, 27 
"Readers or Buyers," 71 
Red Badge of Courage, The, 62 
Red Beach Assault, 58-59, 62-64, 66, 74 
"Red Dust," 109 
Redbook, 84 
Remarque, Eric, 180-181 
Return to Peyton Place, 106 
"Richard Cory," 213 
Richardson, Samuel, 16, 29, 115 
Rinehart, Mary Roberts, 85, 88-89, lo8 
Rise of Silas Lapham, The, 164 
Robert Elsmere, 24, 158-159 
Roberts, Kenneth, 89, 92, 104 
Robinson, E. A., 213 
Rockwell, Norman, 44 
Rogues, Royalty, and Reporters, 158 
Rouse, W. H. D., 158 
Rowson, Susanna, 17 
Russell, Charles Edward, 160 

Salvation Army, 24 

Sandberg, Carl, 85 

Saturday Evening Post, 44, 83-84, 87-89, 

91, 95-96, 98 
Saturday Review, 85, 118 
Scarlet Letter, The, 30, 35-40, 47, 107 
Schulberg, Budd, 118-119 
Scott, Sir Walter, 16, 21-22 
Scribner's Magazine, 81 
Sea-Wolf, The, 132 
Secret Achievement, The, 26 
Sedgwick, Anne, 208 
Serenade, 130 
Seventeen, 84 
Shapiro, Lionel, 181 
Shaw, Irwin, 179-180, 192-196, 198 
Shaw, Loren, 170-171 
Sheldon, Charles, 24, 51 
Sheldon, Walt, 189, 204 
Sheltering Sky, The, 210 
Ship's Company, 181 
Shore Leave, 143, 180 
Sienkiewicz, Henry, 27 
Silver Chalice, The, 43, 47 
Simms, William Gilmore, 17, 21, 30 
Simon Peter, 43 
Sinclair, Upton, 84, 86, 160 
Singular Life, A, 24 
Sister Carrie, 40 
Sixth of June, The, 181 
"Slump in American Writing, The," 28 
Smith, Harrison, 95 
Smith, Joseph L., 157 
Smollett, Tobias (George), 29, 151 
"Snows of Kilimanjaro, The," 210 
Soldier's Pay, 180 
"Song of Solomon," 49, 79, 143 
South Pacific, 181 
Southern Literary Messenger, 21 
Southworth, Mrs. E. D. E. N., 23 
Spillane, Mickey, 31, 134 
Standard Oil Corporation, 160 
Stark, Willie, 146 
Stead, William T., 24 
Stephens, Ann, 23 
Steps Toward Heaven, 19 
Sternberg, Joseph Von, 123 
Sterne, Laurence, 16 
Strange Land, The, 189 
Strether, Lambert, 40 
Sun Also Rises f The, 143, 210 
Sure Hand of God, The, 149, 151 
Swedish Academy, 208 
Swiggett, Howard, 167-168 

224 The Denatured Novel 

Tabor, Ariel, 101 

Tank Destroyers, The, 180, 196-197, 201 

Tarbell, Ida M., 160 

Tarkington, Booth, 40-42, 85, 89, 92, 

98-104, 1 08 
Tasker, George, 56, 58 
Tender Is the Night, 91 
Ten North Frederick, 213 
Thackeray, William Makepeace, 16 
They Bank on Me, 58-59, 63-64 
This Very Earth, 151, 153 
Thomas Young Crowell, 25 
Thompson, Daniel Pierce, 21 
Thoreau, Henry, 143 
Those Devils in Baggy Pants, 181 
Three of a Kind, 129 
Three Soldiers, 192-193, 196 
Ticknor, William, 23 
Tinker, Earl J., 41-42 
Tin Pan Alley, 46 
Tobacco Road, 143, 149 
To Have and Have Not, 123-124 
Tombs, Temples & Ancient Art, 157 
Tragic Ground, 151 
Trouble in July, 151-153 
Troubling of a Star, 189, 204 
Turmoil, 100-101 
Turner, Lana, 126 
Twain, Mark, 134, 143, 151 
20th Century Fox, 106, in 

Ulysses, 116 

United States Census of Manufacturers, 70 
University of Oklahoma Press, 157 
Uris, Leon, x8i, 185 

Vamp, 114-115 
Van Doren, Carl, 95 
Vanrcvel, Tom, 101 
Vichy Government* i a$ 
Vistavision, 186 1.0. 
Vitascope, iio-m 
V-J Day, 178 
Vogue, 84 

Wakeman, Frederic, 143, 180 
Walden, Ty Ty, 154 
Walk in the Sun, A, 62, 180 
Wallace, Lew, 107 
Walling, Don, 170-171, i?4 
Wall Street Journal, 63 
Wanamaker, John, 15 

War and Peace, 32, 181 

Ward, Elizabeth Stuart Phelps, 24 

Ward, Mrs. Humphry, 24 

Warner Brothers, in 

Warner, Susan, 23, 158 

Warren, Robert Penn, 146 

Waterfront, 118 

Wellman, Paul, 158 

West, Mae, 114-115 

West, Nathanael, 147 

West, Rebecca, 88 

Western Electric, in 

Western Printing and Lithography Com- 
pany, 140 

Westward Ho! 18 

Wharton, Edith, 85, 88, 208 

"What is Worth While" Series, 26 

Whipple, E. P., 21-22, 102 

"Why Write It When You Can't Sell It 
to the Pictures?" 118 

Wide, Wide World, The, 23, 158 

Wi eland, 19 

Wilder, Thornton, 31 

"Will the Novel Disappear?" 27 

Wilkins, Mary E., 85 

Williams, Ben Ames, 89, 108 

Williams, Tennessee, 142 

Williams, Wirt, 181, 187 

Wilson, Edmund, 91, 143 

Wilson, Sloan, 167, 171-172, 174 

Wisteria Cottage, 147 

Witchcraft, Magic and Alchemy, 158 

Wolfe, Thomas, 62, 134 

Wolfert, Ira, 179-180 

Woman's Home Companion, 83, 85 

World War I, 166, 180, 183 

World War II, 62-63, 65, 105, in, 135, 
138, 1 66, 178, 180, 184 

Worth and Loesser, 53 

i, 1 86, 198-200, 

[arold Bell, 24, 27, 31 
\ook, The, 86 


Young Lions, The, 180, 192-196, 198 

Zenith, U.S.A., 41 

Zola, Emile, 165 

Zoomies, Subs, and Zeros, 158 

Zukor, Adolph, 112-113